Skip to main content

Full text of "Industrial Evolution"

See other formats


Google 



This is a digital copy of a book that was preserved for generations on Hbrary shelves before it was carefully scanned by Google as part of a project 

to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 

are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 

publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we liave taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 
We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for informing people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liabili^ can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http : //books . google . com/| 



« 



.! 




UnRARy^iSSSg^ OF THE li 








X 






INDUSTRIAL EVOLUTION 



CARL BUCHER 

niiatcr qf nmcal Bammy, UKfutrtOy of Lilpr^ 



TgAHSLATSD FROM THE THIRD GSRMAS BDtTlOS 



S, MORLEY WICKETT, Ph.D. 
X naSciilEcimomy md Statistics. UafumSy ^TtevMb 




NEW YORK 

HENRY HOLT AND COMPANY 
1907 









:L\ ^-.-/'-^■''^^'' •^••^••■^•'■ ' 



Copyright, 1901, 

BT 

HENBT HOLT A CO. 

Published July igox. 



•  



• w 



\K 



o 



\ 



PREFATORY NOTE. 

The writings of Professor Bucher, in their German 
dress, require no introduction to economists. His ad- 
mirable work The Population of Frankfurt in the Fourteenth 
and Fifteenth Centuries, published in 1886, gave him 
immediate celebrity with economic historians, and 
left him without a rival in the field of historical statis- 
tics. In his treatment of economic theory he stands 
midway between the " younger historical school " of 
economists and the psychological Austrians.^ A full list 
of his writings need not be given.^ But I may recall 
his amplified German edition of Laveleye*s Primitive 
Property, his little volume The Insurrections of the Slave 
Labourers, 143-129 B.C., his original and suggestive 
Labour and Rhythm, discussing the relation between the 

^ A few facts and dates regarding Professor Bdcher's career may not 
be uninteresting. Professor Bllcher was born in Prussian Rhineland in 
1847. He completed his undergraduate studies at Bonn and Gdttingen 
(1866-69). His rapid rise in the German scholastic world is evident 
from his academic appointments : special lecturer at Gdttingen (1869- 
72), lecturer at Dortmund (1872-73), at Frankfurt Technical School (1873- 
78), and at Munich (1881) ; Professor of Statistics at Dorpat, Russia 
(1882), of Political Economy and Finance at Basel (1883-90), at Karls- 
ruhe (1890-93), and at Leipsic (1893 to present). From 1878 to the close 
of 1880 he was Industrial and Social Editor of the Frankfurter ZHtung, 

' This may be found in the HandwOrterbuch d. Staatswiss. 

ill 






> -i ^ *i 



IV PREFATORY NOTE. 

physiology and the psychology of labour, his investiga- 
tions into trusts, and his co-editorship of Wagner's Hand- 
book of Political Economy (the section Industry being in 
his charge) as indicating the general direction and scope 
of his researches. The present stimulating volume, which 
in the original bears the title Die Entstehung der Volkswirt- 
schaft (The Rise of National Economy), g^ves the author's 
conclusions on general industrial development. Some- 
what similar ground has been worked over, among recent 
economic publications, alone by Professor Schmoller's 
comprehensive Grundrisz der allgemeinen Volkswirtschafts- 
lehre, Pt. I. But the method of treatment and the results 
of the present work allow it to maintain its unique position. 
Chapters I. and II. outline the prominent features of 
primitive economic life in the tropical zone. These real- 
istic accounts of the " pre-economic stage of industrial evo- 
lution," preceding the dawn of civilization, ably em- 
phasize the kinship of economics and ethnology. In chap- 
ter III. he presents brilliantly and concisely the suggestive 
series of economic developmental stages of household, 
town and national economy, based on the industrial rela- 
tion of producer to consumer; and Chapter IV. offers a 
masterly survey of industrial systems — domestic work, 
wage-work, handicraft, commission work (house indus- 
try), and the factory. With these chapters may be classed 
Chapter V. The Decline of the Handicrafts, The re- 
maining chapters analyze more specifically, from the 
viewpoints of the individual and society, some of the great 
processes of industrial evolution: union and division of 
labour; the intellectual integration of society as effected 
by the press; the formation of social classes; and the fur- 
ther adjustment of labour through internal migrations of 
population. At the same time they enfich economic 
terminology with many telling expressions. 



PREFATORY NOTE. V 

" The worst use of theory is to make men insensible to 
fact," Lord Acton remarked in the opening number of the 
English Historical Review. Our author, with his store 
of minute facts, his keen analysis and his broad and re- 
freshing generalizations, has known how to avoid the 
snare. His historical attitude is indicated by his advice 
that " our young political economists " be sent on jour- 
neys of investigation to the Russians, the Roumanians and 
the southern Slavs rather than to England and America. 
In the following pages, which in their present form I 
trust do not entirely obliterate the pleasing style of the 
original, his attention is, of course, devoted primarily to 
economic rather than to social and other considerations. 

The volume has had in Germany an unusually influ- 
ential circulation, and has recently been translated into 
French, Russian and Bohemian. As the preface notes, 
it has done extensive service as a general introduction to 
" economic thinking." Its use for this purpose, through 
the medium of special transcriptions, has already been re- 
marked at some American universities. The hope may be 
indulged that its merits will now receive wider recognition, 
and in some measure impart to the reader the stimulus 
felt by the writer during a two years' attendance on the 
author's lectures in 1895-97. Editorial annotations, it may 
be. added, have been confined to the narrowest limits. 

In translating it I have had the valuable assistance of 
my colleague. Dr. G. H. Needier, Lecturer on German, 
University College, to whom I wish to express my deep 
obligations. My thanks are also due Professor Biicher for 
his patient answers to the many queries sent over the water 
to him, to Professor Mavor for varied aid during the work 
of revision, and to. Mr. H. H. Langton and Mr. D. 
R. Keys for help in correcting proofs. 



vi PREFATORY NOTE. 

For the convenience of the general reader a short sup- 
plementary bibliography of recent works in English is 
appended. 

S. M. W. 

University of Toronto, 
April 9, 1901. 



Haddoo, Evolution of Art (1895) ; Lloyd Morgan, Animal Life and Intel- 
ligence (1891), Habit and Instinct (1896) ; Keane. Man Past and Present 
(1899) ; Spencer and Gillen, The Native Tribes of Central Australia (1899) ; 
Mackenzie, An Introduction to SocicU Philosophy (2ded., 1895) ; Giddings, 
Principles of Sociology (1896) ; Gumplowicz, Outlines of Sociology (trans. 
1899) ; Loria, Economic Foundations of Society (trans. 1899) ; Ashley, Eco- 
nomic History (2 pts., 1888-92), Surveys Historic astd Economic (1900); 
Gomme, The Village Community (1890) ; Cunningham, Growth of English 
Industry and Commerce (2 vols., 1890-92), Western Civilization (1898) ; 
Booth, Life and Labours of the People (9 vols., 1889-97) ; Mayo-Smith, 
Emigration and Immigration (1890) ; Weber, The Growth of Cities (1899)-; 
Hobson, 7^ Evolution of Modem Capitalism (1S94). 



FROM THE PREFACES TO THE FIRST AND 
SECOND GERMAN EDITIONS. 

(April i8q3 and November 1897.) 

The lectures in this volume were originally delivered 
before audiences that were not composed of specialists 
exclusively. . . . Each lecture is complete in itself; the 
same trains of thought are indeed occasionally repeated, 
but in a different setting. 

Yet it will readily be perceived that the different parts 
have an inner connection, and supplement each other both 
in subject and method. The fundamental idea running 
through them all is expressed in the third. As need 
scarcely be remarked, the lecture is not printed in the sum- 
mary form in which it was delivered. I trust that the gain 
in accuracy and fulness of statement has not been at the 
expense of clearness. 

The lectures are dominated by a uniform conception of 
the orderly nature of economic development, and by a sim- 
ilarity in the method of treating material. In both respects 
this accords with the practice which I have consistently 
followed ever since the inception of my academic activity 
and which during continued scientific work has become 
more and more clearly and firmly established. With the 
present publication I accede to the wish expressed by many 
of my former auditors in the only form at present possible, 
a form of whose insufficiency I myself am fully conscious. 

In preparing a second edition one thing was clear: the 



viii PREFACE. 

book must be expanded in the direction in which it had 
been most effective. At its first appearance I had hoped 
that the little volume might exert some influence upon the 
method of treating scientific problems; and indeed there 
has appeared in recent years quite a series of writings by 
younger authors (some of whom were seemingly wholly 
unacquainted with my book), in which the results of the in- 
vestigations here published are taken into account. This is 
outwardly evidenced by the use of the concepts and the 
technical expressions that I introduced into the litera- 
ture of the subject, as if they were old, familiar, scientific 
furniture. It is perhaps justifiable to infer from this that 
the book has exercised some influence upon academic 
teaching. 

But it seems to have found its chief circulation in the 
wider circles of the educated public, particularly among 
college students, who have used it as a sort of introduc- 
tion to economics, and as a preparation for economic 
thinking. That naturally decided me to keep their wants 
very particularly in view in revising the book. In order 
to avoid misconceptions, however, I wish to state ex- 
pressly that the employment of the book for this purpose 
requires the concurrent use of a good systematic treatise 
on the principles of political economy. 

The better to meet the need of the larger class just men- 
tioned, I have given some of the lectures of the first edition 
a simpler form, expanding them where necessary, and 
eliminating needless detail. Extensive alteration, how- 
ever, has been confined to the lecture on the Organization 
of Work and the Formation of Social Classes. Here uni- 
formity of treatment seemed to recommend a division into 
two chapters (VIII and IX), and such extensive additions 
as would serve to round off each independently of the 
other. The lecture on the Social Organization of the Pop- 



PREFACE, ix 

ulation of Frankfurt in the Middle Ages has been omitted 
because it disturbed the greater uniformity aimed at for 
the complete work, and because, as a purely historical ac- 
count, it is better suited to a collection of sketches in social 
and economic history, for which opportunity may perhaps 
be found later. 

On the other hand three unpublished lectures have been 
added (Chaps. I, V, and VII). The first of these deals 
with the pre-economic period, and is intended to furnish 
the substructure for the system of economic stages which 
is developed in the third chapter. Its main features were 
sketched as early as 1885, in a lecture I delivered at the 
University of Basel on the beginnings of social history. 
. . . The second agrees in the major part of its matter, 
and also to a large extent in its form, with the report upon 
handicrafts that I presented at the last general meeting 
of the Social Science Club in Cologne. It seems advisable 
to give it a place in order to afford the reader at one point 
at least an insight into the great changes that are in prog- 
ress in the field of modern industrial life. The third [en- 
titled " Union of Labour and Labour in Common "] is an 
attempt to lay before a wider circle of readers, in the form 
in which I finally presented it to my university classes, a 
chapter from the theory of labour to which I have given 
considerable attention. 

All the lectures in this volume, both old and new, were 
originally sections of university lectures. Every lecturer 
knows what a wonderful compilation his note-book is, how 
from semester to semester certain portions must be re- 
moved and reconstructed, how some parts are never 
approached without an inward struggle, and how finally 
the remaining difficulties are surmounted and the whole 
given a form satisfactory alike to teacher and students. 
To the lecture-room first of all belong the fruits of 



X PREFACE. 

the scientific labours of the German university professor; 
but he also naturally wishes to submit what he has labori- 
ously accomplished to the critical judgment of specialists; 
and for my part I feel in such cases the further need of test- 
ing the maturity of my conceptions by seeing whether they 
can be made intelligibly acceptable to a wider range of 
readers. So that all the lectures that have been taken over 
from the first edition were actually delivered before a more 
popular audience, while Chapters I and VII are essays in 
the same style. In the extent of their subject-matter, how- 
ever, they all reach far beyond what can be offered directly 
to students in a university lecture. 

In conclusion I may allude to two attacks that have 
been delivered by historians against some parts of the third 
and fourth lectures. The blame surely does not rest with 
me if these gentlemen have failed to perceive that this 
work treats of economic theory, not of economic history. 
He who, in the outline of a period of development extend- 
ing over thousands of years, expects a minute and ex- 
haustive presentation of the actual condition of any par- 
ticular people and century, need blame only himself if he 
is disappointed. In the first edition I expressed myself 
clearly enough, I think, regarding the logical character of 
the economic stages. In the present edition I have taken 
occasion, however, to give the passages in question such a 
form that in future they cannot with good intentions be 
misunderstood. . . . Though for the central idea of my 
theory of development it is altogether immaterial whether 
I have in every particular characterized the economy of 
the Greeks and Romans correctly or not, or whether the 
guild handicraft of the Middle Ages was chiefly wage-work 
or chiefly independent hand-work (" price-work "). 



PREFACE TO THE THIRD GERMAN EDITION. 

In the present edition, bearing in mind the way in which 
this book came into existence and has since expanded, I 
felt strongly impelled to mark my appreciation of the 
recognition it has gained, as indicated by several editions 
and by translations into French, English, Russian, and 
Bohemian, by preparing additional chapters to remove a 
number of gaps still noticeable in the last issue. If I have 
not yielded to the temptation, it is chiefly for the reason 
that a larger bulk would necessarily prejudice the wider 
circulation of the volume. 

The most disturbing want has been met by the insertion 
of a new chapter on the economic life of primitive peoples 
(Chapter II). The chapter differs from the more detailed 
one in volume 3 of the Yearbook of the Gehe Stiftung 
in its more summary form and in the addition of some not 
unimportant facts. 

All the other chapters have been carefully revised and 
many slight improvements made. More extensive alter- 
ations are confined to Chapters I, III, VII, and VIII. 

May the book in its present form satisfy its old friends 
and add new ones to the number I 

Carl Bucher. 
LupziG, Oct. 15, 190a 



CONTENTS 



ffAGB 

PUFATORT Non. Hi 

From the P&kfacx3 to First akd Second German Editions... vii 

Preface to Third German Edition xi 

Chapter I. — Primitive Economic Conditions i 

II. — The Economic Life of Primitive Peoples 41 

III. — The Rise of National Economy 83 

IV. — A Historical Survey of Industrial Systems.... 150 

V. — The Decline of the Handicrafts 185 

VI. — The Genesis of Journalism 215 

VII. — Union of Labour and Labour in Common 244 

Vin. — Division of Labour 282 

IX.— Organization of Work and the Formation of 

Social Classes 315 

X. — Internal Migrations of Population and the 

Growth of Towns Considered Historically 345 

Index 387 



CHAPTER I. 

PRIMITIVE ECONOMIC CONDITIONS. 

All scientific investigation of industry starts with the 
assumption that man has a peculiar " economic nature " 
belonging to no other living creature. From this eco- 
nomic nature a principle is supposed to spring, which con- 
trols all his actions that are directed to the satisfaction of 
his wants. This is the economic principle, the fundamen- 
tal principle of economic activity. This principle reveals 
itself in man's endeavour always and everywhere to attain 
the highest possible satisfaction with the least possible sac- 
rifice (labour) — the " principle of least sacrifice." 

According to this view all man's economic actions are 
actions directed toward an end and guided by considera- 
tions of profit. Whether or not the final impulse to eco- 
nomic labour is to be sought in the instincts of man (the 
instinct of self-preservation and of self-interest), the satis- 
faction of these instincts is always the result of a series of 
successive mental operations. Man estimates the extent 
of the discomfort that would arise from the non-satisfac- 
tion of a want felt by him; he measures the discomfort that 
the labour necessary to meet the want can cause him; he 
compares the discomforts with each other, and resolves to 
undertake the labour only when the accompanying sacrifice 



2 PRJMITiyE ECONOMIC CONDITIONS. 

is less than the sacrifice of remaining unsatisfied. More- 
over, upon undertaking the labour he again chooses the 
least burdensome among the various possible methods of 
procedure, and thus has a further series of considerations, 
estimations, comparisons, and judgments to enter upon. 

In fact the whole science of political economy proceeds 
on the assumption that economic actions have behind them 
a rational motive and call into play the higher mental facul- 
ties; and it has evolved a kind of psychology of labour, by 
means of which it seeks to explain those actions in their 
typical progress. Economic activity is, therefore, some- 
thing especially human; indeed the question whether the 
lower animals display similar activity, seems never to have 
been broached. The economic nature 'of man is something 
absolute, inseparable from the very character of man.^ 

Yet even among civilized mankind, from whose manifold 
activity the principle of economy has been deduced, indi- 
cations are not wanting to show that the economic nature 
must be characteristic of different individuals in different 
degree. Between the industrious and the indolent, the 
provident and the improvident, the sparing and the spend- 
thrift, there are innumerable gradations; and if we only 
observe the conduct of the child towards his possessions, 
we are easily convinced that the " economic nature " must 
be acquired anew by each human being, and that it is a 
result of education and custom, in which individuals differ 
no less than in their whole physical and mental develop- 
ment. 

Having once reached this point, we shall scarcely be 

*"The elements of economic character are firmly rooted in the 
physical and intellectual organization of man, and change just as little 
as his outward character does, at least in the periods which come within 
the scope of the history of mankind." — Wagner, Grundlegung d. polit. 
Oekonomie (3. Aufl.), I, p. 82. [As for animal sociology, it can hardly 
be said to have advanced as yet beyond animal psychology. — Ed.] 



PRlMlTiyE ECONOMIC CONDITIONS, 3 

able to postpone the question, whether indeed that " eco- 
nomic nature " does not, for mankind in general, signify 
something acquired rather than something given by na- 
ture; and whether we must not assume at the beginning 
of human evolution a period of purely instinctive satisfac- 
tion of wants reaching over many thousands of years, such 
as we are accustomed to take for granted in the case of the 
lower animal. 

The answer to this question can be gained only by pro- 
ceeding inductively. The picture of primitive man that we 
make for ourselves must be not an imaginary one — no 
Robinson Crusoe story such as is so often encountered in 
the deductions of the " classical " economists. Its Hues 
must all be drawn from reality; they must show us the 
actuality of the assumed conditions under which uncivil- 
ized man lives and the impulses under which he is con- 
ceived to act and later also think. Civilized man has al- 
ways had a great inclination to read his conceptions and 
feelings into the mind of primitive man; but he has only 
a limited capacity for understanding the latter's undevel- 
oped mental life and for interpreting, as it were, his nature. 

To be sure, aboriginal man in actual existence can no- 
where now be found. Great as is the number of primitive 
peoples that have gradually come within our ken, none of 
them stands any longer at the lowest stage of savagery; 
all show traces of the first step in civilization, for all know 
the use of fire. 

Many writers, it is true, have imagined, under the stim- 
ulus of evolutionist theories, that they had succeeded, 
now here, now there, in discovering populations pre- 
serving the original animal state down to the present. 
As late a writer as Sir John Lubbock is inclined to 
deny to several tribes of the South Sea Islands a 
knowledge of fire. O. Peschel has been at pains to 



4 PRIMITiyE ECONOMIC CONDITIONS. 

prove that the instances adduced by Lubbock are in- 
correct ; ^ and with him we may regard as valid the asser- 
tion that upon the whole earth the tribe that has not made 
use of fire is yet to be found. Even the prehistoric cave 
discoveries, which show us men of the Ice Age along with 
the bear, the aurochs, and the reindeer, show traces of the 
use of fire. Fire indeed is a powerful influence in the direc- 
^ tion of civilization. It enlarges man's sphere of suste- 
nance, teaches him to harden the points of the wooden 
arrows and spears, to hollow out the tree, and to frighten 
away the wild beasts. 

Other investigators have imagined they have discovered 
human beings who lived together in small groups in trees, 
had fruits for food, and used only stones and cudgels as 
weapons and instruments, after the fashion of the higher 
apes. Frederick Engels * is of the opinion that by this as- 
sumption alone can we explain the continued existence of 
man alongside the great beasts of prey. Lippert,who inves- 
tigates the case more carefully,* finds, it is true, that in the 
myth of the Egyptians the tree plays a certain role as an 
abode of spirits; but he is prudent enough not to conclude 
from this that the ancestors of the Egyptians dwelt in 
trees, — more prudent than the philologist Lazarus Geiger, 
who discovered a relic of tree-dwelling in the hammock 
used by the South American Indians. It is true that in 
Sumatra, Luzon, New Guinea, the Solomon Islands, and 

* Races of Man (New York, 1888), pp. 137 ff. I know, indeed, that 
the American writer Teale (quoted by Lippert, KuHurgeschichte der 
Menschheit, I, p. 52) has contradicted him in one instance. Mundt- 
Lauff has also, according to Peschel in " Natur " for the year 1879, p. 
478, denied the use of cooked food by the Negritos in the Philippines, 
but his assertions again have been refuted by A. Schadenberg in the 
Ztschr. f. Ethnologie, XII (1880), pp. 143-4. [No ethnologist would 
now claim fireless tribes as known in actual existence. — Ed.] 

 Der Ursprung der Familie, des Privateigentums u, des Staats, p. 7. 

* As above, pp. 67 flF. 



PRIMITIVE ECONOMIC CONDITIONS. 5 

among the Gaberi negroes in Central Africa, huts have 
been found built in between the branches of large trees; '^ 
and the same is reported of individual forest tribes of South 
America.® But so far as these products of primitive archi- 
tecture are not mere temporary protective structures that 
are supplemented by permanent dwelling-places upon the 
ground, they are by no means the most unfinished of hab- 
itations, and the peoples using them prove by many kinds 
of implements, utensils, and domestic animals, and some 
of them even by the agriculture they carry on, that they no 
longer stand at the first dawn of civilization. 

After what has been said there can be no object in 
searching out uncivilized peoples and beginning with a 
description of them, after the example of Klemm who 
opens his General History of Civilization with the Forest 
Indians of Brazil, although it is not to be denied that these 
stand at a very low cultural stage indeed. In the same 
connection^ other investigators cite tribes standing at no 
higher stage: the Bushmen in South Africa, the Batuas 
in the Congo basin, the Veddahs in Ceylon, the Mincopies 
in the Andaman Islands, the Negritos in the Philippines, 
the Australians of the continent, the now extinct Tas- 
manians, the Kubus in Sumatra, and the Tierra-del-Fue- 
gians. To whom to adjudge the palm for savagery might 
be difficult to decide. O. PescheP finds individual ele- 
ments of civilization among them all, even among the 
Botokudos, whom he himself considers still nearest the 
primitive state. 

The assumption of such a primitive condition, in which, 

* Nachtigal, Sahara u. Sudan, II, pp. 628 ff. Finsch, Samoafahrten, 
pp. 271 f. Ratzel, VdlkerkundCj I, pp. loi, 105, 245, 586; II, p. 83. 
[Its different arrangement precludes citation from the admirable Eng- 
lish edition: The History of Mankind, 2 vols., London, 1896-97. — Ed.] 

•Waitz, Anthropologic d. Naturvolker, III, p. 393. 

* Races of Man, pp. 149 ff. 



6 PRlMITiyE ECONOMIC CONDITIONS. 

armed with no other resources than are at the command of 
the lower animal, man has to join in the struggle for exist- 
ence, is one of the necessary expedients of all sciences that 
aim at a history of man's development; and this is true of 
sociology and especially of political economy. We must, 
however, abandon the attempt to exemplify the primitive 
condition by any definite people. On the other hand, there 
is more prospect of scientific results in an endeavour to 
collect the common characteristics of the human beings 
standing lowest in the scale, in order, by starting ' with 
them, to arrive at a picture of the beginnings of economic 
life and the formation of society. But in this it is by no 
means necessary to confine ourselves to the above-men- 
tioned representatives of the lowest manner of life; for 
every delimitation of that kind would challenge objections 
and contract the field of vision. Moreover the various ele- 
ments of mental culture and material civilization are by 
no means so mutually dependent that all must necessarily 
develop at an equal pace, and thus we find among almost 
all primitive peoples characteristics that can have sprung 
only from the most ancient mode of Hfe. The collection 
of these characteristics, and their combination into a 
typical picture, must, however, be our first task. 

Hitherto the process has usually been made too simple 
by deriving the characteristics of primitive man from civil- 
ized economic man. The many wants of man in a state of 
nature, so it has been argued, demanded for their satisfac- 
tion exertions beyond the capacity of the individual; pro- 
tection from wild beasts or from the unchained elements 
could likewise be attained only by the labour of many. 
Accordingly writers have spoken of a collective carrying- 
on of the struggle for existence, and thus have had " prim- 
itive society " and a sort of communistic economy com- 
plete. 



PRIMlTiyE ECONOMIC CONDITIONS. 7 

But man has undoubtedly existed through immeasur- 
able periods of time without labouring. If so disposed, 
one can find plenty of districts upon the earth where 
the sago-palm, the plantain-tree, the breadfruit-tree, the 
cocoa- and date-palm, still allow him to live with a mini- 
mum of exertion. It is in such districts that tradition is 
fondest of placing paradise, the original home of mankind; 
and even modern research cannot dispense with the as- 
sumption that mankind was at first bound to such regions 
of natural existence and only by further development be- 
came capable of bringing the whole earth into subjection. 

Of unions into organized society we find, moreover, 
hardly a trace among the lowest races accessible to our 
observation. In little groups ® like herds of animals they 
roam about in search of food, find a resting-place for the 
night in caves, beneath a tree, behind a screen of brush- 
wood erected in a few minutes to shelter them from the 
wind, or often in a mere hollow scooped in the ground, 
and nourish themselves chiefly with fruits and roots, 
though all kinds of animal food, even down to snails, mag- 
gots, grasshoppers, and ants are eaten also. The men as 
a rule are armed simply with bow and arrow or with a 
throwing-stick ; the chief implement of the women is the 
digging-stick, a pointed piece of wood, which they use in 
searching for roots. Shy when they come in contact with 
members of a higher race, often malicious and cruel, they 
lead a restless life, in which the body, it is true, attains the 
maximum of agility and dexterity, but in which technical 
skill advances only with extreme sluggishness and one- 
sidedness. The majority of peoples of this type know 
nothing whatever of pottery and the working of metals. 
Even of wood, bast, stone, and bone they make but limited 

*Comp. on this point £. GrossCi Die Formen der FamUie und die 
Formen der Wirtschaft, p. 37. 



S PRIMmyE ECONOMIC CONDITIONS. 

use, and this use leads in no way to a stock of utensils and 
tools, which indeed it would be impossible to carry about, 
because of their nomadic life, bearing as it does the char- 
acter of one continuous search for food.® 

• In order to supplement this general account by a few details I will 
here introduce a portion of the description of the Negritos in the 
Philippines published in the work by A. Schadenberg, cited above. 
I give for the most part his own words: — The women among the 
Etas bear easily and quickly. Until able to walk, the child is carried 
by the mother, generally on the left thigh, in which case it assumes 
a sort of riding posture; or upon the back, as soon as it is able to 
hold itself on. The mother nurses it for about two years. At about 
the tenth year puberty comes; the Negrito youth is then tattooed, 
and from the moment when this decoration of his body is finished he 
is independent. He accordingly looks about him for a mate, who has 
in the majority of cases been selected for him beforehand and who, 
if possible, belongs to the same " family." The members of a " family/' 
which generally numbers twenty to thirty persons, are under the con- 
trol of an elective chieftain, who decides upon the camping-places 
and the time for breaking up. The family life is patriarchal in char- 
acter. The father has unlimited power over the members of his 
family; he can chastise them and even barter away his children; the 
woman occupies a subordinate position and is tceated as a chattel. 
The Negritos carry on bartering with the Tagalas; in this way they 
get supplies, chiefly of iron, in exchange for honey and wax. By 
means of the iron thus acquired they prepare part of their weapons, 
which consist of hunting-knives, arrows, bows, and spears. The 
Negritos are also very clever at throwing stones, in which they are 
greatly assisted by their keenness of vision. A stone in the hand of 
an otherwise unarmed Negrito is thus an offensive and defensive 
weapon not to be despised. Their clothing is very scant, hardly more 
than a breech-clout. Domestic utensils for permanent use are 
scarcely found at all among the Etas; sometimes a clay vessel gdt 
by barter with the Malays, and as a rule a piece of bamboo from 
three to four metres in length for holding drinking-water. Their 
toes are prehensile, and are of great assistance to them in climbing. 
In the matter of food they are not fastidious; it is both animal and 
vegetable in character — roots, honey, frogs, deer, wild-boar, etc. A 
Spanish ecclesiastic describes them as follows: "The pure Aitos or 
Negritos lead a secluded life; they have no fixed dwelling-place and 
do not build huts. The father, the mother, and the children are all 
provided with arrows, each having his own, and they set out together 



PRIMITiyE ECONOMIC CONDITIONS. 9 

These peoples have been designated " lower nomads " ; 
but it can scarcely be proven that actual hunting forms 
their chief means of sustenance. They all make use of 
vegetable food as far as it is at all obtainable, and with 
those who live in a warmer climate this food seems to pre- 
dominate. Stores of such fruits and roots they do not 
gather, though a region of plentiful supplies attracts a 
greater number of members of the tribe, as a rich feeding- 
ground draws together many lower animals; when it is 
exhausted they scatter again. And similarly as to the 
mollusks and insects which they consume; each individ- 
ual at once swallows what he finds; joint household life 
is as little known as is a house. It is only when a large 
animal has been killed or found dead (the fondness for 
meat in a state of decay is widespread) that the whole 
group assemble,^ ^ and each devours as much as he can; 

upon the hunt. When they kill a deer or a boar, they halt at the 
spot where the animal has fallen, scoop a hole in the ground, place 
the animal in it and then build a fire. Each one takes the piece of 
the animal that suits his taste best and roasts it at the fire. And so 
they go on eating until they have filled their bellies, and when thus 
satiated they sleep on the earth which they have hollowed out, as 
pigs do when they have gorged themselves. When they awake 
they go through the same operation, and so on until all the meat is 
devoured; then they set out upon the hunt again." They observe no 
fixed times for sleeping and eating, but follow necessity in both cases. 
They age early; at forty or fifty the mountain Negritos are decrepit, 
white-haired, bent old men. — Compare further the descriptions of 
the Botokudos by Ehrenreich, Ztschr. f. Ethnol., XIX, pp. i ff ; of 
the Bororos by K. v. d. Steinen, Unter d. Naturvolk. Central-Brasil.f 
pp. 358 ff. ; of the Bushmen by Fritsch, as above, pp. 418 ff. ; of the 
Veddahs by P. and F. Sarasin, Die Weddas von Ceylon; of the Aus- 
tralians by Brentano, Ztschr. f. Sozial- u. Wirtschaftsgeschichte, I, 

pp. 133-4. 
"From the custom prevalent among some of the lower tribes of 

proclaiming the finding of food by means of loud calls, Lippert, as 

above, I, p. 246, concludes that " the consideration due to the family " 

is expressed in this way. In this connection it is to be observed that 



10 PRIMITIVE ECONOMIC CONDITIONS, 

but the method of hunting these animals strongly resem- 
bles the procedure of the wild beast stealing upon its prey. 
With their imperfect weapons these peoples are hardly 
ever in a position to kill an animal instantly; the chief 
task of the hunter consists in pursuing the wounded game 
until it sinks down exhausted.^ ^ 

Regarding the constitution of the family among peo- 
ples of this class, there has been much discussion. Of late 
the opinion seems to be gaining ground that there exists 
among them a fellowship between man and wife that ex- 
tends beyond a mere sexual relationship and is of life-long 
duration; while upon the other hand it cannot be denied 
that in case of a scarcity of food those loose groups easily 
split up, or at least individual members detach themselves 
from them. Only between mother and child is the rela- 
tionship particularly close. The mother must always take 
her little one aiong with her on the march; and for that 
purpose she usually fastens it in some way on her back, a 
custom that is very general among all primitive peoples, 
even where they have gone over to agriculture. For sev- 
eral years the child must be nurtured at the breast or from 
the mother's mouth, but it soon acquires skill in procuring 
its food independently, and often separates itself from the 
community in its eighth or tenth year. 

All the tribes involved in our survey belong to the 
smaller races of mankind, and in bodily condition give the 

many animals (for example, our domestic fowl) have the same cus- 
tom. True, he lays stress upon the fact that no one thinks of collect- 
ing stores of provisions. Therefore, we are, further, not justified in 
agreeing to the proposal that has recently been made from several 
sides to designate these peoples as gatherers of stores (Sammler). 

"Comp. G. Fritsch, Die Eingehorenen Sud-Afrikas, pp. 324, 425: 
Pogge, Im Reiche d. Muata Jamwo, pp. 328-9; Wissmann, Im Innem 
AfrikaSy pp. 260, 341; Martius, Zur Ethnographie Amerikas, zumal 
Brasiliens, pp. 665 ff. 



PRIMITIVE ECONOMIC CONDITIONS. il 

impression of backward, stunted growth. We are not on 
that account, however, justified in regarding them as de- 
generate race fragments. The evidence rather goes to 
show that the more advanced races owe their higher 
physical development merely to the regular and more 
plentiful supply of food which agriculture and cattle- 
raising for centuries past have placed within their reach, 
while the peoples here in question have always remained 
at the same stage. Subject to all the vicissitudes of the 
weather and the fortune of the chase, they revel at times 
in abundance and devour incredible masses of food; still 
oftener, however, they suffer bitter want, and their only 
article of clothing, the breech-clout, is for them really the 
" hunger-strap " (" Schmachtriemen ") of German story, 
which they tighten up in order to alleviate the pangs of 
gnawing hunger.^^ 

" How from this stage of primitive existence the path 
leads upward is manifest in countless typical examples 
furnished by ethnology. In addition to the collection of 
wild fruits and roots, the woman takes over the cultiva- 
tion of food-plants. This she carries on at first with the 
customary digging-stick, later with a short-handled hoe. 
The man continues his hunting and fishing, which, in rich 
hunting-grounds and with more perfect weapons, he can 
make so productive that they furnish the greater part of 
his food. At times he supplements these by cattle-raising. 
In the procuring of food each sex has its sharply defined 
sphere of duties to which with advancing technical skill 
there are added in each case various industrial arts, which 
as a rule retain their connection with the original produc- 
tion and occupation. Among advanced primitive peoples 

"On the Bushmen comp. Fritsch, as above, p. 405; on the Aus- 
tralians, Peschel, Races of Man, p. 332; on the Botokudos, Ehrenreich 
in the Ztschr. f. Ethnol., XIX (1887). p. 27, 



12 PRIMITIVE ECONOMIC CONDITIONS. 

all economic activity may be traced back to combinations 
of these elements; but in its details such labour is entirely 
dependent upon local natural conditions. We should 
therefore not be justified in any attempt to construct stages 
of development intended to hold equally good for negroes, 
Papuans, Polynesians, and Indians. 

But wherever we can observe it, the method in which 
primitive peoples satisfy their wants reminds us continu- 
ally in many of its features of the instinctive doings of the 
lower animals. Everywhere their existence is still far from 
settled, and even the unsubstantial huts they erect are for 
the majority only temporary structures which, 4jowever 
much they vary from place to place and from tribe to tribe, 
always remain true to a type, and remind us of the nests of 
birds, which are deserted as soon as the brood is fledged. 

When Lippert finds the fundamental and controlling 
impulse to cultural development in material foresight, he 
undoubtedly makes an advance upon earlier investigators; 
but the phrase itself is not happily chosen. It is utterly 
impossible to speak of foresight, in the sense of providing 
for the f.uture, in connection with primitive peoples. 
^ Primitive man does not think of the future; he does not 
think at all; he only wills, that is, he wills to preserve his 
existence. The instinct of self-preservation and self-grati- 
'^ fication is the prime agent of development, compared with 

which even the sexual instinct takes quite a secondary 
place. 

Wherever it has been possible for Europeans to observe 
men in primitive conditions for any length of time they 
tell us of the incomparable dulness and mental inertness 
which strike the beholder; of their indifference to the sub- 
limest phenomena of nature, their complete lack of interest 
in everything that lies outside the individual self. The 
savage is willing to eat, sleep, and, where necessary, to pro- 



PRlMITiyE ECONOMIC CONDITIONS. i$ 

tect himself against the greatest inclemencies of the 
weather: this is his whole aim in life. 

It is therefore entirely false and contrary to numer- 
ous well-accredited observations when Peschel straightway 
ascribes to savages a peculiar wealth of fanciful imag- 
inings of a religious nature, and thinks that the closer the 
approach to the condition of nature the greater the range 
of belief. He evidently assumes that the course of the sun 
and the other phenomena of the heavens must be infinitely 
more impressive and stimulative of active thought to the 
primitive than to the civilized man. But that is by no 
means the case. Both among the Indians of Brazil and 
among the negroes, when travellers have asked about 
these things, the response has been that people never 
thought about them; and Herbert Spencer ^^ has collected 
an abundance of examples which show that the lower races 
do not manifest any interest even in the most novel phe- 
nomena. The Patagonians, for example, displayed utter 
indifference when they were made to look into a mirror; 
and Dampier reports that the Australians whom he had 
taken on board of his ship paid attention to nothing but 
what they got to eat. Burton** calls the East Africans 
" Men who can think, but who, absorbed in providing for 
their bodily wants, hate the trouble of thinking. His [the 
East African's] mind, limited to the object seen, heard, 
and felt, will not, and apparently can not, escape from the 
circle of sense, nor will it occupy itself with aught but the 
present. Thus he is cut off from the pleasures of memory, 
and the world of fancy is altogether unknown to him." ^^/ 

The same force, then, that impels the lower animal, the 

** Principles of Sociology ^ vol. I, 8§ 45-6. 

** The Lake Regions of Central Africa (New York, i860), p. 489. 

"Comp. the similar opinion of the missionary Cranz, Historic von 
Gronland (Frankfurt, 1780), p. 163, and Lubbock, Origin of Civilisation 
(4th ed.), pp. 516-7. 



14 PRIMITiyE ECONOMIC CONDITIONS. 

instinct for preserving its existence, is also the dominant 
instinctive impulse of primitive man. This instinct is lim- 

^ ited in scope to the single individual; in respect of time, 
to the moment at which the want is felt. In other words: 
the savage thinks only of himself ^ and thinks only of tlie 
present. What lies beyond that is as good as closed to his 
mental vision. When, therefore, many observers reproach 
him with a boundless egoism, hardness of heart towards his 
fellows, greed, thievishness, inertness, carelessness with re- 
gard to the future, and forgetfulness, it means that sym- 
pathy, memory, and reasoning power are still entirely un- 
developed. Nevertheless it will be wise for us to make 
these very characteristics our starting-point, in order to 
comprehend the relation of primitive man to the external 
world. 

In the first place, as concerns the egoism of the savage 
and his hardness of heart towards his nearest relations, this 
is a natural consequence of the restless nomadic life in 
which each individual cares only for himself. It shows 
itself most prominently in the extraordinarily widespread 

^ custom of infanticide, which extremely few primitive peo- 
ples are entirely free ivom}^ The children impede the 
horde on the march and in the search for food. Thereiii 
lies the chief reason for their removal. Once become a 
custom, infanticide lives on in later stages of civilization; 
indisputable traces of it have been found not merely among 
the primitive peoples of Asia, Africa, America, Australia, 
and Polynesia, but even among the Arabs, the Romans, 
and the Greeks. 

To infanticide is universally ascribed the exceptionally 
slow increase of the uncivilized races. But this is also de- 
pendent upon their short lives, and long lactation pe- 

**Comp. Lippert, II, pp. 201 ff.; Ratzel, Volkerkunde, I, pp. 108, 154, 
252, 277, 306, 338, 425. 



PRlMlTiyE ECONOMIC CONDITIONS, 15 

riods, during which, as is well known, conception rarely 
occurs, and this it is which forms the chief cause of their 
protracted tarrying at the same stage of civilization. That 
the natural bond between parents and children is nowhere 
very firm is seen also in the extremely common custom of 
adoption.*'' It is even said, for example, that in the " fam- 
ilies " of the Mincopies the children of other parents are in 
the majority. It is significant that between their own and 
their adopted children they make, as a rule, not the slight- 
est difference. Adoption may have arisen from the substi- 
tution of child-exposure for infanticide. If the natural 
mother was not in a position to take the new-born child 
along with her, perhaps another woman who was childless 
could, and thus the life of the child was saved. 

Recent ethnographers have been at great pains to prove 
that the strength of maternal love is a feature common 
to all stages of civilization. It is, indeed, a matter of 
regret to us that we find wanting in our own species a 
feeling that exhibits itself in such a pleasing way among 
many families of lower animals. But there have been too 
many observations showing that among the lower races 
the mere care for one's own existence outweighs all other 
mental emotions, in fact that beside it nothing else is of the 
least importance. All observers are amazed and even in- 
dignant at the indifference with which children, when once 
they can shift for themselves, separate from their blood- 
relations.*® Yet we have here only the reverse side of that 

" Comp. Lubbock, Origin of Civilisation, pp. 95-6. 

" Comp. the striking example in Ratzel, Volkerkunde, I, p. 677, of a 
boy of Ticrra del Fuego who, when taken on board a European ship, 
did not show the slightest grief over the separation, while his parents 
were delighted to get a few necklaces and some biscuits in return for 
him. The selling of children and women into slavery does not occur 
in Africa alone. Martius, as above, p. 123. Comp. Post, Afr, Juris' 
prudeng, I, p. 94- 



1 6 PRlMITiyE ECONOMIC CONDITIONS. 

hardness of heart which " enables husbands to refuse food 
to their wives, and fathers, to deny it to their hungering 
children, when they themselves would but feast upon it." 

This same trait of unbounded selfishness is manifest in 
the regardlessness with which many primitive peoples 
leave behind them on the march, or expose in solitary 
places, the sick and the aged who might be an impediment 
to the vigorous.^® This trait has often been interpreted 
as a sign of superstition, as due to the fear of evil powers 
to whom the illnesses are ascribed. And in fact in the case 
of tribes that have become settled and whose means of sub- 
sistence would admit of the care of the sick, appearances 
favour such an explanation. But at the same time it is 
forgotten that customs, once firmly rooted, perpetuate 
themselves with great persistence, even when the causes 
that gave rise to them have long since passed away. 

From exposure to intentional killing is only a short step. 
Indeed we find even among peoples on a higher plane of 
civilization that old age is deplored as a state of extreme 
joylessness. Barbarism had no aflfection between relatives 
to alleviate this condition, but it had it in its power to 
shorten it; and so, along with exposure, we find the bury- 
ing, or the killing, or even the devouring of the aged and 
sick, as numberless examples from Herodotus down to 
modem times attest. Indeed primitive man was able to 
look upon the solemn performance of this horrible act as 
a behest of piety .^^ 

When we see how this unbroken nomadic life forced 
man to devote his whole activity to the securing of food. 



1* 



Lippert, as above, pp. 229 ff., has treated the subject so exhaus- 
tively that I may refrain from citing examples. Comp. also Fritsch, 
PP- 116, 334» 351; Waitz, as above, II, p. 401. 

" Comp. the examples cited by Lippert, p, 232, and Martius, as 
above, p. 126. Also Ehrenreich, Beitrdge s. Vblkerkunde BrasU.f pp. 
69-70; Waitz, as above, I, p. i8g—Introd. to Anthropol. (trans.), p. 161. 



PRIMITIVE ECONOMIC CONDITIONS. 17 

and forbade the concurrent development of those feelings 
which we regard as the most natural, and how it even sue- 

• 

ceeded in giving the appearance of religious duty to what 
we consider the most abominable crime, we begin to con- 
ceive how loose must have been the personal bond that 
held together those little roving groups of human beings. 
Sexual intercourse could not grow to be such a bond; for 
what we call love was entirely wanting in it.^^ Domestic 
life, the conception of property and labour in common 
were as good as non-existent. These could originate only 
when the circle of wants advanced beyond the mere food 
requirement. But this took a much longer time than the 
majority are willing to admit. The needs of primitive peo- 
ples with regard to clothing and house-shelter are most 
markedly of an altogether secondary nature. 

Turning now to the no less common characteristic of 
improvidence, we must certainly at first glance be struck 
with astonishment. One would think that hunger, which 
often brings such great torture to the savage, would of 
itself have been sufficient to induce him to store up for 
future use the food that at times he possesses in super- 
abundance. But the observations that have been made 
all indicate that he never thinks of that. " They are not 
accustomed," says Heckewelder ^^ of the North American 
Indians, " to laying in stores of provisions except some 
Indian corn, dry beans, and a few other articles. Hence 
they are sometimes reduced to great straits, and are not 
seldom in absolute want of the necessaries of life, especially 

"The many writers who write nowadays about the family pay 
altogether too little attention to this point, to which prominence has 
justly been given by Lubbock, as above, pp. 72 ff. In the same way 
they overlooked the connection between the family and the economy 
of the home. [Comp. p. 10 above. — Ed.] 

** Heckewelder, Indian Nations, etc. New edition (Phil., 1881), pp. 
198* 212 (Memoirs of Hist. Soc. of Penn., vol. 12). 



1 8 PRIMITIVE ECONOMIC CONDITIONS, 

in the time of war." And of the South American tribes 
another observer reports: ^^ " It is contrary to their na- 
ture to be in possession of food-supplies for longer than 
one day at most." With many negro tribes it is looked 
upon as improper to store up food for future need, which 
belief, it is true, they base upon the superstition that the 
fragments left over may attract spirits.^* 

Where these peoples, through the short-sighted greed 
of gain of Europeans, are placed in possession of modern 
weap>ons, they usually work incredible havoc with the game 
in their hunting-grounds. The extermination of the 
boundless buffalo herds of North America is well known. 
" The greatest quantities of meat were left lying unused 
in the thickets," only for the natives in winter-time, when 
deep snow prevented hunting, to fall a prey to awful 
hunger, in which even the bark of trees and the roots of 
grass were not despised. And to-day the natives of Africa, 
in districts where they carry on a profitable trade with 
Europeans, are ruthlessly destroying the sources of their 
incomes, the elephant and the caoutchouc-tree. 

Even among the more advanced tribes and individuals 
this characteristic does not fail. " When the carriers re- 
ceived fresh rations," relates P. Pogge,^® " I am certain 
that they lived better for the first few days than I did. The 
best goats and fowl were devoured. If I gave them rations 
for a fortnight, the rule was to consume them in riotous 
living during the first three or four days, only afterwards 
either to steal from the supply-trains, to beg from me, or 
to go hungry." In Wadai everything that remains over 
from the sultan's table is buried,^^ and at the sacrificial 

"Appun, Unter den Tropen^ p. 365. 
•• Lippert, as above, I, pp. 39-40. 

"As above, p. 14: comp. p. 6. Also Wissraann, Wolf, etc., Im 
Innem Afrikas, p. 29. 
" Nachtigal, Sahara u. Sudan, III, p. 230. 



PRIMITiyE ECONOMIC CONDITIONS. 19 

feasts of the Indians the guests were obliged to eat up their 
meat and bread clean. " Overloading of the stomach and 
vomiting are not unusual on such occasions." ^® 

Closely connected with this waste of supplies is the use 
that primitive man makes of his time. It is entirely errone- 
ous, though customary, to imagine that primitive people 
are particularly expert in measuring time by the position 
of the sun. They do not measure time at all, and accord- 
ingly do not make divisions in it. No primitive people 
observe fixed meal-times, according to which civilized 
man regulates his time for work.^® Even such' a relatively 
advanced tribe as the Bedouins has no conception of time. 
They eat when they are hungry. Livingstone in one place 
calls Africa " the blissful region where time is absolutely 
of no account and where men may sit down and rest them- 
selves when they are tired." *® " Even the most trivial work, 
though it is urgently necessary, is postponed by the negro 
to as late a date as possible. The native dreams away 
the day in laziness and idleness, although he knows quite 
well that for the night he needs his draught of water 
and his log of wood; nevertheless until sundown he will 
certainly not disturb himself, and only then, or perhaps 
not before darkness, will he finally procure himself the 
necessaries.'* '* 

In these words we have touched upon the reproach of 
inertia to which primitive man is universally subject.'^ 

*• Heckewelder, as above, p. 213. [Dr. Bucher, quoting from the 
German translation, has evidently mistaken the meaning of the pas- 
sage cited. The vomiting and fasting referred to by Heckewelder are 
preparatory to the ceremonies, the vomiting being self-induced. — Ed.] 

"Comp. W. Wundt, Ethics^ I (London, 1897), pp. 171-2. 

** Expedit. to the Zambesi (New York, 1866), p. 104. 

"W. Junker's Travels in Africa-, comp. Eng. trans., II, p. 168. 

" For details see my work Arbeit u. Rhythmus (2d ed., Leipzig, 
1899). 



ao PRIMITiyE ECONOMIC CONDITIONS. 

What has here appeared to observers as laziness is again 
lack of forethought, living for the moment. Why should 
the savage exert himself when once his wants are satisfied, 
particularly when he is no longer hungry? This does not 
imply that he is inactive. With his wretched facilities the 
individual often performs on the whole as much work as 
the individual civilized man; but he does not perform it 
regularly, nor in ordered succession, but by fits and starts, 
when necessity forces him to it, or a feeling of exaltation 
takes possession of him, and even then not as a serious 
duty of life, but rather in a playful fashion. 

In general, primitive man follows only the prompting of 
the moment; his conduct is purely impulsive, mere reflex 
action, so to speak. The nearer his wants and their satis- 
faction lie together, the better he feels. Primitive man is 
a child; he thinks not of the future, nor of the past; he 
forgets easily, each new impression blots out its predeces- 
sor. All the sufferings of life, which he has to experience 
so often, can scarcely cloud for a moment his naturally 
cheerful temperament. "Of the New Caledonians, Fijians, 
Tahitians, and New Zealanders we read that they are al- 
ways laughing and joking. Throughout Africa the negro 
has the same trait, and of other races, in other lands, the 
descriptions of various travellers are: 'full of fun and merri- 
ment,' ' full of life and spirits,' * merry and talkative,' ' sky- 
larking in all ways,' ' boisterous gaiety,' * laughing immod- 
erately at trifles.' " ^^ 

It is significant, as has often been remarked, that natives 
of Africa lose their cheerfulness when they have been for 
some time in the service of Europeans, and become sullen 
and morose. The explanation of Fritsch ** is that servants 

" Spencer, as above, § 34. Considerable material also in his Descrip' 
five Sociology in the chapter on " Moral Sentiments." 
•• As above, p. 56. 



PRIMITIVE ECONOMIC CONDITIONS. 21 

of this kind gradually acquire from their masters the habit 
of troubling themselves about things to come, and that 
their temperament cannot endure engrossment in such 
cares.*" 

Such a hand-to-mouth existence cannot be burdened 
with conceptions of value, which always presupposes an act 
of judgment, an estimation of the future. It is well known 
how in America and Africa the natives often sold their land 
to foreign colonists for a gaudy trifle, a few glass beads of 
no value according to our economic standards; and even 
to-day the negro, though he stands no longer at the lowest 
stage, is in many instances ready to give away any piece 
of his property, no matter how important it may be for his 
existence, if he is offered some glittering bauble that hap- 
pens to catch his eye.*® On the other hand his covetous- 
ness knows no bounds, and it is a constant complaint of 
travellers that amid all the hospitality shown them they 
are simply plundered, because every village chieftain de- 
sires to be presented with everything he sees.*" Here, 
again, is that naive egoism in its complete regardlessness 
of self and others, that unbounded covetousness which has 
nothing in common with the love of gain of economic 
man. The impression of the moment is ever the sole con- 
trolling force; what is further removed is not thought of. 
Primitive man is incapable, it would seem, of entertaining 

* Reference may also be made to the not infrequent examples of 
savages who had been brought up under civilized conditions volun- 
tarily returning to their tribes and to the complete savagery of their 
people, Comp. Peschel, as above, pp. 152-3; Fritsch, as above, p. 
423; K. E. Jung in Pctermanns Mitth., XXIV (1878), p. 67. 

■* Comp. Fritsch, as above, pp. 305-6. 

"Says Burton, as above, p. 499: "In trading with [the African] 
. . . all display of wealth must be avoided. A man who would pur- 
chase the smallest article avoids showing anything beyond its equiva- 
lent." 



2 2 PRIMITIVE ECONOMIC CONDITIONS. • 

two thoughts at the same time and of weighing the one 
against the other; he is always possessed by one alone and 
follows it with startling consistency. 

The collection of experiences, the transmission of knowl- 
edge, is therefore exceedingly difficult, and therein lies the 
chief reason why such peoples can remain at the same 
stage for thousands of years without showing any appre- 
ciable advance. The acquisition of the first elements of 
civilization is often conceived as an easy matter; it is im- 
agined that every invention, every advance in house-con- 
struction, in the art of clothing, in the use of tools, that an 
individual makes, must pass over as an imperishable treas- 
ure into the common possession of the tribe and there 
continue to* fructify. The invention of pottery, the taming 
of domestic animals, the smelting of iron ore have even 
been made the beginnings of entirely new epochs of civil- 
ization. 

Yet how imperfectly does such a conception appreciate 
the conditions under which primitive man lives! We may 
indeed assume that he possesses a peculiar fondness for the 
stone axe which with endless exertion he has formed in the 
course perhaps of a whole year, and that it seems to him 
like a part of his own being; ^® but it is a mistake to think 
that the precious possession will now pass to children 
and children's children, and thus constitute the basis for 
further advance. However certain it is that in connection 
with such things the first notions of "mine and thine" are 
developed, yet many are the observations indicating that 
these conceptions do not go beyond the individual, and that 
they perish with him. The possession passes into the grave 
with the possessor, whose personal equipment it formed in 
life. That is a custom which is met with in all parts of the 

"Comp. Arbeit u. Rhythmus (2d ed.), p. 16. 



PRIMITIVE ECONOMIC CONDITIONS. 23 

earth, and of which many peoples have preserved traces 
even down into civilized times.®® 

In the first place it is prevalent among all American peo- 
ples to such an extent that the survivors are often left in 
extreme need. The aborigines of California, who are 
among the lowest people of their race, place along with 
the dead all the weapons and utensils which he had made 
use of in life. " It is a curious paraphernalia," says an ob- 
server, " that follows the )A^intum into the grave: knives, 
forks, vinegar-jars, empty whiskey-bottles, preserve-jars, 
bows, arrows, etc. ; and if the dead has been an industrious 
housewife, a few baskets of acorns are scattered over 
as well." " At the grave of the Tehuelche " (Patagonia), 
runs another account, " all his horses, dogs, and other ani- 
mals are killed, and his poncho, his ornaments, his bolas 
and implements of every kind are brought together into a 
heap and burned." And of a third and still lower tribe, 
the Bororo in Brazil, a recent very reliable observer says: ^^ 
" A great loss befalls a family when one of its members 
dies. For everything that the dead man used is burned, 
thrown into the river, or packed in with the corpse in order 
that the departed spirit may have no occasion whatever to 
return. The cabin is then completely gutted. But the 
surviving members of the family receive fresh presents; 
bows and arrows are made for them, and when a jaguar 
is killed, the skin is given to the brother of the last deceased 
woman or to the uncle of the last deceased man." 

Among the Bagobos in southern Mindanao (one of the 

■• Comp. in general, Andree, Ethnogr. Parailelen u, Vcrgleiche (Stutt- 
gart, 1878), pp. 26-7; Schurtz, Grundriss einer Entsteh. Gesch. d. 
Geldes (Weimar, 1898), pp. 56 ff.; Pauckow, Ztschr. d. Ges. f. Erdkunde 
zu Berlin, XXXI, pp. 172-3. 

*• K. von den Stein en, Unter den NcUurvdlkern BrasUiens (2d ed.), 
p. 389. Comp. also Ehrenreich, Beitrdge zur V'dlkerkunde BrasUiens, 
pp. 30, ^; Heckewelder, as above, pp. 270-1, 274-5. 



24 PRIMITIVE ECONOMIC CONDITIONS. 

Philippines) the dead man is buried in his best clothes, and 
with a slave, who is killed for the purpose. " The cooking- 
utensils that the deceased used during his lifetime are filled 
with rice and placed, along with his betel-sacks, upon the 
grave; his other things are left in the house untouched. 
On penalty of death no one may henceforth enter either 
the house or the burial-place; and it is equally forbidden 
to cut from the trees surrounding the house. The house 
itself is allowed to go to ruin." *^ 

In Australia and Africa the custom is very common for 
all the stores of the deceased to be eaten up by the assem- 
bly of the mourners; in other parts the utensils are de- 
stroyed, while the food is thrown away. Many negro 
tribes bjiry the dead in the hut in which he lived, and leave 
the dwelling, now deserted by the survivors, to decay; 
others destroy the hut."*^ If a chieftain dies, the whole vil- 
lage migrates; and this is true even of the principal towns 
of the more important kingdoms, such as those of the 
Muata-Yamwo and the Kasembe. In the Lunda kingdom 
the old royal Kipanga is burned down, and a new pro- 
visional one at once erected, for which the newly chosen 
ruler has to kindle a fresh fire by rubbing together pieces 
of wood, as it is not permissible to use the old fire any 
longer.. The principal or residence town changes its loca- 
tion with each new ruler.f * Among the ancient Peruvians, 



41 



Schadenberg in the Ztschr. f. Ethnol, XVII (1885), pp. 12-13. The 
same thing is found in Halamahera, p. 83; and among the hill tribes 
of India, Jellinghaus in the same review, III, pp. 372, 374. 

** Examples may be found in M. Buchner, Kameruny p. 28; Fritsch, 
as above, p. 535; Bastian, LoangokustCy I, p. 164; Livingstone, as 
above, p. 158. From Australia: Parkinson, Im Bismarck-Archipel, pp. 
102-3; Ztschr. f. Ethnol., XXI, p. 23; Kubary, Ethnogr. Beitrdge zur 
Kenntnis d. Karolin, Inselgruppe u. Nachharschaft (Berlin, 1885), pp. 
70-71, note. 

^ Pogge, as above, pp. 228, 234. Livingstone in Petermanns Mitth., 
XXI (i875)» P- 104. 



PRJMITiyE ECONOMIC CONDITIONS, 25 

as well, the conception prevailed that with each new Inca 
the world, so to speak, began anew. The palaces of the 
dead Inca, with all their stores of wealth, were closed for 
ever; the ruler for the time being never made use of the 
treasures that his ancestors had amassed. 

Though we see from this that the origin and preserva- 
tion of the first elements of civilization among primitive 
peoples were attended with the greatest difficulties, and 
that the possibility of rising to better conditions of exist- 
ence and higher modes of life could not even be conceived 
by them, yet it must not be forgotten that the observations 
that have been sifted and presented here have been taken 
from peoples of very varied character and different cultural 
stages. To raise himself of his own strength to the plane 
of the Tongan or Tahitian, the Australian of the continent 
would probably have required many thousands of years; 
and a similar gulf separates the Bushman from the Congo 
negro and Wanyamweza. But this very fact, it seems to 
me, speaks for the persistence of the presumptive psychic 
conditions under which the satisfaction of the wants of un- 
civilized man is accomplished; and we are undoubtedly 
justified in tracing back the whole circle of this class of 
conceptions to a condition that must have prevailed 
among mankind for aeons before tribes and peoples could 
have originated. 

From all that we know of it, this condition means ex- 
actly the opposite of " economy." For economy implies 
always a community of men rendered possible by the pos- 
session of property; it is a taking counsel, a caring not 
only for the moment but also for the future, a careful divi- 
sion and suitable bestowal of time; economy means work, 
valuation of things, regulation of their use, accumulation 
of wealth, transmission of the achievements of civilization 
from generation to generation. And even among the 



a6 PRIMITIVE ECONOMIC CONDITIONS. 

higher primitive peoples we have found all this widely 
wanting; among the lower races we have hardly met 
with its faint beginnings. Let us strike out of the 
life of the Bushman or of the Veddah the use of fire, 
and bow and arrow, and nothing remains but a life 
made up of the individual search for food. Each indi- 
vidual has to rely entirely upon himself for his sustenance. 
Naked and unarmed he roams with his fellows, like cer- 
tain species of wild animals, through a limited stretch of 
territory, and uses his feet for holding and climbing as 
dexterously as he uses his hands.^^ All, male and female, 
devour raw what they catch with their hands or dig out of 
the ground with their nails: smaller animals, roots, and 
fruits. Now they unite in little bands or larger herds; 
now they separate again, according to the richness of the 
pasturage or hunting-ground. But these unions do not 
develop into communities, nor do they lighten the exist- 
ence of the individual. 

This picture may not have many charms for him who 
shares the civilization of the present; but we are simply 
forced by the material empirically gathered together so to 
construct it. Nor are any of its lines imaginative. We 
have eliminated from the life of the lowest tribes only what 
admittedly belongs to civilization — the use of weapons and 
of fire. Though we were obliged to admit that even 
among the higher primitive peoples there is exceeding 
much that is non-economic, that at all events the con- 
scious application of the economic principle forms with 
them rather the exception than the rule, we shall not be 
able in the case of the so-called '^ lower nomads " and their 
predecessors just sketched to make use of the notion of 
economy at all. With them we have to fix a pre-economic 

**R. Andree, Der Fuss als Greif organ, in his Ethnogr. Parallekn «. 
Vergl. (New Series), pp. 228 ff. 



PRIMITIVE ECONOMIC CONDITIONS. 27 

Stage of development, that is not yet economy. As every 
child must have his name, we will call this the stage of in- 
dividual search for food. 

How economic activity was evolved from the individual 
search for food can to-day hardly be imagined. The 
thought may suggest itself that the turning-point must 
be where production with a more distant end in view takes 
the place of the mere seizing of the gifts of nature for im- 
mediate enjoyment, and where work, as the intelligent 
application of physical power, replaces the instinctive 
activity of the bodily organs. Very little would be gained, 
however, by this purely theoretical distinction. Labour 
among primitive peoples is something very ill-defined. 
The further we follow it back, the more closely it ap- 
proaches in form and substance to play. 

In all probability there are instincts similar to those that 
are found among the more intelligent of the lower animals, 
that impel man to extend his activities beyond the mere 
search for food, especially the instinct for imitating and 
for experimenting.*® The taming of domestic animals, for 
example, begins not with the useful animals, but with such 
species as man keeps merely for amusement or the worship 
of gods. Industrial activity seems everywhere to start 
with the painting of the body, tattooing, piercing or other- 
wise disfiguring separate parts of the body, and grad- 
ually to advance to the production of ornaments, masks, 
drawing on bark, petrograms, and similar play-products. 
In these things there is everywhere displayed a peculiar 
tendency to imitate the animals which the savage meets with 
in his immediate surroundings, and which he looks upon 
as his equals. The partly prehistoric rock drawings and 
carvings of the Bushmen, the Indians, and the Australians 

*• Coxnp. K. Groos, Dii SpieU d, Tiere (Jena, 1896). 



28 PRIMITIVE ECONOMIC CONDITIONS. 

represent chiefly animals and men; ^® pottery, wood-carv- 
ing, and even wicker-work begin with the production of 
animal forms.*^ Even when the advance is made to the 
construction of objects of daily use (pots, stools, etc.), the 
animal figure is retained with remarkable regularity; *^^ 
and lastly, in the dances of primitive peoples, the imita- 
tion of the motions and the cries of animals plays the prin- 
cipal part.*^^ All regularly sustained activity finally takes 
on a rhythmic form and becomes fused with music and 
song in an indivisible whole.*^* 

It is accordingly in play that technical skill is developed, 
and it turns to the useful only very gradually.** The or- 
der of progression hitherto accepted must therefore be 
just reversed; play is older than work, art older than pro- 



•Andree, Ethnogr. ParalleUn u, Vergleiche, pp. 258-299. Ehrenreich, 
as above, pp. 46-7. 

•"Comp. the interesting accounts by K. v. d, Steinen, as above, 
pp. 231 ff., and particularly pp. 241 fT. 

" Abundant examples are afforded by every ethnographic picture- 
collection. The incredulous are invited to take a glance at the fol- 
lowing works: J. Boas, The Central Eskimo (Washington, 1888); 
Sixth Annual Report of the Bureau of Ethnology to the Secretary 
of the Smithsonian Institution, 1884-5; Ethnogr. Beschrijving van de 
West' en Noordkust van Nederlandsch Niew Guinea, by F. S. A. de Clercq 
and J. D. E. Schmeltz (Leiden, 1893); Joest, Ethnogr. aus Guyana 
(Suppl. to vol. 5 of the Intern. Archiv fiir Ethnogr.) ; and again Von 
den Steinen, as above, pp. 261 flF. Comp. also Fritsch, as above, p. 73; 
Schweinfurth, The Heart of Africa (3d ed., London, 1878), I, pp. 129- 
130; and Grosse, Die Anfdnge d. Kunst, Chaps. VI and VII. 

" Grosse, as above, pp. 208-9. 

" Reference must here again be made to my work on Arbeit u. 
' Rhythmus. 

" " On examination of the primitive tools [of the Papuans] . . . 
we see that there is not a single one which does not bear testimony by 
some little design or ornament to the good taste of its maker, not an 
article which does not show some trifling accessories surpassing mere 
utility and present in it solely for beauty's sake." — Semon, In the 
Austral. Bush, p. 400. 



PRIMITIVE ECONOMIC CONDITIONS. 29 

duction for use.^' Even when among the higher primi- 
tive peoples the two elements begin to separate from each 
other, the dance still precedes or follows every more im- 
portant work (war-, hunting-, harvest-dance), and song ac- 
companies work. 

Just as economy, the further we have traced it back- 
wards in the development of peoples, has during our in- 
quiry assumed more and more the form of non-economy, so 
work also has finally resolved itself into its opposite (Nicht- 
arbeit). And we should probably have the same experi- 
ence with all the more important phenomena of economy, 
if we were to continue our inquiries regarding them. One 
thing alone appears permanent — consumption. Wants 
man always had, and wants must be satisfied. But even 
our wants, considered from an economic point of view, 
exist only in very small part naturally; it is only in the 
matter of bodily nourishment that our consumption is a 
necessity of nature; all else is the product of civilization, 
the result of the free creative activity of the human mind. 
Without this activity man would always have remained a 
root-digging, fruit-eating animal. 

Under these circumstances we must forego the attempt 
to fix upon some definite point at which simple search for 
food ceases and economy begins. In the history of human 
civilization there are no turning-points; here everything 
g^ows and decays as with the plant; the fixed or station- 
ary is only an abstraction which we need in order to make 
visible to our dim eye the wonders of nature and humanity. 
Indeed economy itself, like all else, is subject to constant 
changes. When it first presents itself in history, it appears 
as a form of communal life based upon material posses- 

"* [The general ethnographic aspect of art is, of course, another and 
wider problem than the one involved in this conclusion with regard 
to " first things."— Ed.] 



30 PRIMmVE ECONOMIC CONDITIONS. 

sions, guided by definite rules of conduct, and closely con- 
nected with the personal and moral community of life of 
the family.*^® It was under this form that it was seen by 
the people who first fixed its characteristics in language. 
Landlord is still in Middle High German synonymous 
with husband (Wirt, Ehemann), landlady is wife (Wirtin, 
Ehefrau), and the word economy, derived from the Greek, 
is formed in a similar way. 

We may therefore assume the existence of economy as 
certain where we find co-dwelling communities that pro- 
cure and utilize things adapted to their needs according to 
the dictates of the economic principle. Such a condition 
is certainly fulfilled by the higher primitive peoples, even 
though their carrying out of the economic principle always 
remains incomplete. But there is nevertheless much that 
still recalls the pre-economic period of individual search 
for food; economy has still, so to speak, gaps in» various 
places. 

Among all peoples of a lower cultural stage the dis- 
tribution of labour between the two sexes is firmly fixed 
by custom, although difference of natural aptitude seems 
by no means to have been the sole determining factor. 
At least it cannot be maintained that in all cases the lighter 
share of the work fell to the weaker sex. While in the 
normal domestic economy of civilized nations we have a 
cross-section, so to speak, which assigns to the man the 
productive work and to the woman the superintendence 
of consumption, the economy of these peoples seems to be 
divided in longitudinal section. Both sexes take part in 

" E. Grosse has in his recent book entitled Die Formen d. Familie 
u. die Formen d, Wirtschaft (Leipzig, 1896) investigated the connection 
of family with economic forms. lA doing so he has, for the economic 
side of the work, adhered to the altogether external classification into 
nomadic, pastoral, and agricultural, but has scarcely devoted due at- 
tention to the inner economic life, particularly to that of the household. 



PRIMlTiyE ECONOMIC CONDITIONS. 31 

production, and frequently each has a particular depart- 
ment of the consumption for itself. It is particularly signif- 
icant in this connection that upon the woman devolves, as a 
rule, the procuring and preparing of the vegetable foods 
and for the most part also the building of the hut, while 
hunting and the working up of the products of the chase 
fait to the man. If cattle are kept, the tending of the ani- 
mals, the erection of inclosures for them, the milking, etc., 
are the business of the men. This division is often so 
sharply defined that we might almost speak of a cleavage 
of the family economy into a purely male and a purely 
female economy. 

In an interesting account of the useful plants of the 
Brazilian Shingu tribes, K. von den Steinen^* describes 
the outcome of the earlier development of these tribes in 
the following words: 

"Man followed the chase, and in the mean time woman invented 
agriculture. Here, as throughout Brazil, the women have exclusive 
charge not only of the preparation of manioc in the house but also 
of its cultivation. They clear the ground of weeds with pointed sticks, 
place in position the stakes with which the manioc is planted, and 
fetch each day what they need, carrying it home in heavily laden 
wicker baskets. . . . The man is more courageous and skilful; to 
him belong the chase and the use of weapons. Where, then, hunt- 
ing and fishing still play an important part, the woman must attend, 
as far as a division of labour takes place at all, to the procuring, trans- 
porting, and preparing of the food. This division has a result that is 
not sufficiently appreciated, namely, that the woman in her field of 
labour acquires special knowledge just as the man does in his. This 
must necessarily hold true for each lower or higher stage. The 
counterpart of the Indian woman cultivating her manioc with skill 
and intelligence is already found in the purely nomadic state. The 
wife of the Bororo went into the forest armed with a pointed stick to 
search for roots and tubers; during wanderings through the camp- 
ing-ground, or whenever a company of Indians changed its locality, 
such hunting fell to the woman, while the man tracked the game; she 
climbed the tree for cocoanuts, and carried heavy loads of them 

■* Unter d. Naturvblk. Central-BrasU., pp. 206 ff. 



32 PRlMITiyE ECONOMIC CONDITIONS. 

laboriously home. And though the Indian woman was the slave of 
her husband, this relation was certainly not to her advantage in re- 
gard to the division of the fish and meat; she still had to depend 
upon the supply of vegetables that she could gather for herself. On 
the Shingu the men made ready the roast and broiled fish and meat; 
the women baked the beijus (manioc cakes), prepared the warm 
drinks, cooked the fruits, and roasted the cocoanuts. What other 
meaning could this division into animal cooking for the men and 
vegetable cooking for the women have than that each of the sexes had 
still kept to its original sphere? " 

If we add that the men also made the weapons for the 
chase, and that hunting and fishing were the source whence 
they drew all their implements for cutting, scraping, polish- 
ing, piercing, tracing, and carving, while the women pro- 
duced the pottery for cooking,*^^ we have for each sex a 
naturally defined sphere of production upon which all their 
activity is independently expended. But this is not all. The 
consumption is also in one chief particular distinct: there 
are no common meals in the family. Each individual eats 
apart from the rest, and it is looked upon as improper to 
partake of food in the presence of others.*^® 

Similar characteristics of an individual economy are also 
found among the North American Indians, who had al- 
ready arrived at a fully developed domestic economy. 
While they know nothing whatever of a special ownership 

" As above, pp. 197 flf., 217 ff., 318. 

"• Von den Steinen, as above, p. 69, and Ehrenreich, Beitrdge z. Vol- 
kerkunde BrasU., p. 17: " Etiquette among the Karaya demands that 
each one shall eat by himself apart from the others." Eating alone 
is almost suggestive of the animal. The savage acts like the dog, 
which grows cross when his meal is disturbed. Among the natives of 
Borneo " the men usually feed alone, attended on by the women, and 
always wash their mouths out when they have finished eating. They 
are very particular about being called away from their meals, and it 
takes a great deal to make a man set about doing anything before he 
has concluded his repast; to such an extent is this practice observed, 
that it is considered wrong to attack even an enemy whilst he is eat- 
ing, but the moment he has finished it is legitimate and proper to fall 
upon him." Hose, in Joum. of Anthropol. Inst, p. 160. 



PRIMITIVE ECONOMIC CONDITIONS. 33 

of the soil, " there is nothing in an Indian's house or family 
without its particular owner. Every individual knows 
what belongs to him, from the horse or cow down to the 
dog, cat, kitten, and little chicken. Parents make presents 
to their children, and they in return to their parents. A 
father will sometimes ask his wife or one of his children 
for the loan of a horse to go a-hunting. For a litter of 
kittens or brood of chickens there are often as many dif- 
ferent owners as there are individual animals. In purchas- 
ing a hen with her brood one frequently has to deal for it 
with several children." ^'^ 

" In cases where the Indians permitted polygamy, it was 
customary for a special hut to be erected for each wife; 
among tribes dwelling in common-houses each wife had at 
least her special fire.*^® 

" The same characteristics are presented by the economy 
of the Polynesians and Micronesians, except that here fish- 
ing and the raising of smaller live stock take the place of 
hunting. In New Pomerania the various duties are strictly 
divided between the men and boys on the one hand, and 
the women and girls on the other.*^® To the male portion 
of the population falls the labour of making and keep- 
ing in repair the weapons and the fishing-gear, especially 
the fishing-nets and the ropes necessary for them, of cast- 
ing the nets in the sea and caring for them daily, of build- 
ing canoes, erecting huts, and, in the wooded districts, of 
felling trees and clearing away the roots for the Jaying out 
of new plantations, as well as of protecting them by en- 
closures against wild boars." ®^ 

" Heckewelder, as above, p. 158. 
"Waitz, AfUhropologie, III, p. 109. 
■•Parkinson, as above, pp. 113, 122. 

* These labours preparatory to the cultivation of the land are still 
often performed by the women. Parkinson, p. 118. 



34 PRIMITIVE ECONOMIC CONDITIONS, 

Besides having to care for their little children it de- 
volves upon the women to prepare the food, to dig and 
cultivate the ground, to raise and gamer the produce of the 
field, and to carry the heavily laden baskets to market- 
places miles away. 

" In certain kinds of work both men and women take 
part. To these belong the twisting of the strong basten 
cordage of which the fishing-nets are woven, the plaiting 
of baskets with finely cut strips of rattan and padanus- 
leaves, the weaving of a very rough and coarse stuff called 
ntal, made from the bark of the broussonetia-tree, in which 
the women wrap their infants to protect them from the 
cold." 

This latter is very significant: we have here to do with 
arrangements for the transformation of materials, such as 
could not have existed in the period of individual search 
for food. 

Separate preparation of food for men and women, and 
separate meals are also met with in the South Sea regions. 
In Fiji the men prepare such kinds of food as can be 
cooked out of doors by means of heated stones. " This is 
confined to-day to the roasting of swine's flesh; formerly 
the preparation of human flesh was also reserved for the 
men." ®^ In the Palau Islands the cooking of the taro and 
the preparing of the sweetened foods fell to the lot of the 
women, the preparation of the meats to the men.®^ In 
most parts of Oceania " it is neither permissible for women 
and men to eat in common, nor for the men to eat what 
the women have prepared. Eating with another from the 
same dish seems to be avoided with almost equal scrupu- 
lousness." ®* 

•* Bassler, Siidsee-Bilder, pp. 226-7. 

•• Kubary, as above, p. 173. 

•*Ratzel, Vblkerkunde, I, p. 240. 

•* Separate meals for men and women; comp. Stanley, How I Found 



PRIMITIVE ECONOMIC CONDITJONS. 35 

The economy of many negro tribes shows a like arrange- 
ment; a sharp division of the production and of many parts 
of the consumption according to sex, indeed even the ex- 
tension of this distinction to the sphere of barter. As P. 
Pogge,**^ one of our most reliable observers, says concisely 
of the Congo negfroes: " The woman has her own circle 
of duties independent of that of her husband." And in the 
description of the Bashilangas he observes: ®® " No mem- 
ber of the family troubles himself about another at meal- 
times; while some eat the others come and go just as it 
suits them; but the women and the smaller children gen- 
erally eat together." And finally he reports further re- 
garding the Lundas: " Under ordinary conditions, when 
a caravan has pitched its camp in a village, the women of 
the place are accustomed to bring vegetables and fowl into 
the camp for sale, while goats, pigs, and sheep are usually 
sold only by the men.®''^ It is similarly related by L. Wolf ®® 
that in the market at Ibaushi all the agricultural products 
and materials, mats, and pottery are sold by the women, 
and only goats and wine by the men. Each sex is thus 
possessor of its special product of labour, and disposes of 
it independently.*® 

The division of the labour of production between the two 
sexes in Africa varies in detail from tribe to tribe; as a rule, 

Livingstone (New York, 1887), p. 550; Nachtigal, Sahara «. Sudan, I, 
p. 664. 

"/w Reiche d. Muata Jamwo, p. 40. 

"*Wissmann, Unter deutsch. Flagge quer durch Afrika, p. 387. /m- 
Reiche d, Muata Jamwo, pp. 178, 231. 

" Im Reiche d. Muata Jamwo, p. 29. 

**Wissmann, etc., Im Innern Afrikas, p. 249. Comp. Livingstone, 
Exped. to the Zambesi, pp. 122, 577; Paulitschke, Ethnographie Nordost^ 
Afrikas, I, p. 314. 

* Of the family and the domestic establishment of the Wanyamwezis, 
who assuredly no longer stand upon a low plane, Burton gives the 
following picture : " Children are suckled till the end of the second 
year. . . . After the fourth summer the boy begins to learn archery. 



/ 



36 PRIMITIVE ECONOMIC CONDITIONS. 

however, agriculture and the preparation of all the vegeta- 
ble foods are also assigned here to the woman, and hunting, 
cattle-raising, tanning, and weaving to the man."^^ This 
arrangement is often supported by superstitious usages. 
In Uganda the milking of the cows falls exclusively to the 
men; a woman is never permitted to touch the udder of 
a cow.*^^ In the Lunda territory, again, no man is allowed 
to take part in the extraction of oil from the ground-nut, 
as his presence is thought to frustrate the success of the 
operationJ^ As a rule the carriers whom Europeans en- 



... As soon as [he] can walk he tends the flocks; after the age of 
ten he drives the cattle to pasture, and, considering himself independ- 
ent of his father, he plants a tobacco-plot and aspires to build a hut 
for himself. Unmarried girls live in the father's house until puberty; 
after that period the spinsters of the village, who usually number from 
seven to a dozen, assemble together and build for themselves at a 
distance from their homes a hut where they can receive their friends 
without parental interference. . , . Marriage takes place when the 
youth can afford to pay the price for a wife. [The price] varies, ac- 
cording to circumstances, from one to ten cows. The wife is so far 
the property of the husband that he can claim damages from the adul- 
terer, but may not sell her except when in difficulties. . . . Polygamy 
is the rule with the wealthy. There is little community of interests, 
and apparently a lack of family aflfection in these tribes. The hus- 
band, when returning from the coast laden with cloth, will refuse a 
single shukkah to his wife; and the wife, succeeding to an inheritance, 
will abandon her husband to. starvation. The man takes charge of the 
cattle, goats, sheep, and poultry; the woman has power over the 
grain and the vegetables; and each must grow tobacco, having little 
hope of borrowing from the other. . . . The sexes do not eat to- 
gether: even the boys would disdain to be seen sitting at meat with 
their mothers. The men feed either in their cottages or, more gen- 
erally, in the iwanza [public house, one being set apart for each sex]." 
(As above, pp. 295-8; comp. pp. 493-4-) 

^•Comp. especially Fritsch, as above, pp. /pflf., 183, 229, 325; Living- 
stone, as above, pp. 77 ^ 118, 311-12. An extended treatment now in 
H. Schurtz, Das afr. Gewerbe (Leipzig, 1900), pp. 7 flf. 

^ Emin Bey, in Petermanns Mitth., XXV (1879), P- 392. 

"Wissmann, Wolf, etc., Im Innem Afrikas, p. 63. 



PRIMITIVE ECONOMIC CONDITIONS. 37 

gage refuse to do women's work; Livingstone '^^ even 
reports a case of famine among the men in a certain dis- 
trict because no women were there to grind the com they 
had on hand. The separation of the two sexes in the prep- 
aration and consumption of food is often made still more 
rigid by regulations ^* of a semi-religious character, for- 
bidding the women the use of certain kinds of meat, which 
are thus reserved for the men alone."^* 

Everywhere among primitive peoples the children become 
independent very early in youth and desert the society of 
their parents. They often live then for some years in special 
common-houses, of which there are others for married 
men. These common-houses for the men-folk grouped ac- 
cording to age, and frequently ' also for the unmarried 
women grouped in the same way, are found very widely 
distributed in Africa and America, and especially in 
Oceania. They serve as common places of meeting, work, 
and amusement, and as sleeping-places for the younger 
people, and are used also for lodging strangers. They 
naturally form a further obstacle to the development of a 
common household economy based upon the family, for 
each family is generally subdivided into different parts 

"As above, pp. l88, 565. Similarly among the Indians; comp. 
Waitz, as above, fll, p. 100. 

'* More frequent still in Polynesia. Comp. Andree, Ethnogr, Paral- 
lelen «. Vergleiche, pp. 114 flf. 

" For a peculiar further development of this economy comp. 
Schweinfurth, Sahara u. Sudan, III, pp. 162, 244, 249. In some places 
the separation of the spheres of activity of the two sexes extends even 
to their intellectual life. Among several Caribic tribes the women 
and the men have different names for many things, whence it was 
inferred that there existed distinct languages for men and for women. 
More recently this phenomenon is supposed to have its foundation in 
the difference in social position of the sexes and in the sharp division 
between their two spheres of employment. Comp. Sapper, Intern. 
Archiv. f. Ethnogr., X, pp. 56 f. 



38 PRlMlTiyE ECONOMIC CONDITIONS. 

with separate dwellings. In Yap, one of the Caroline 
Islands, for instance, we find besides the febays, or sleep- 
ing-houses of the unmarried, a principal house for each 
family which the father of the family uses, and also a dwell- 
ing-house for each wife; finally, " the preparation of food 
in the dwelling-house is forbidden and is transferred to a 
separate hut for each member of the family, which serves as 
a fire-cabin or cooking-house." ''^ A similar arrangement 
prevails in Malekula in the New Hebrides.'''^ Further than 
this, economic individualism can hardly be carried. 

It may be asserted as a general rule for primitive peo- 
ples practising polygamy that each wife has her own hut.*^® 
Among the Zulus they go so far as to build a separate 
hut for almost every adult member of the household, — 
one for the husband, one for his mother, one for each of 
his wives and other adult members of his family. These 
huts all stand in a semicircle about the enclosed cattle- 
kraal in such a way that the man's dwelling is in the centre. 
Of course it is to be remembered that a hut of this kind 
can be constructed in a few hours. 

Thus we see that everywhere, even among the more 
developed primitive peoples, there is still wanting much of 
that unified exclusiveness of domestic life with which the 
civilized peoples of Europe, from all we know of them, 
first appeared in history. Everywhere wide clefts still 
gape, and the individual preserves an economic independ- 
ence that strikes us with its strangeness. However much 

''* Kubary, Ethnogr. Beitrdge s. Kennt. d. Karol-Archip. (Leiden), p. 39. 

"Journal of the Anthropol. Inst, of Gr. Br., XXIII (1894), p. 381. 

"This is the case, to mention only a few instances, in the Antilles: 
Starcke, The Primitive Family (New York, 1889), p. 40; in Min- 
danao: Schadenberg, Ztschr. f. Ethnol., XVII, p. 12; among the 
Bakuba: Wissmann, Im Innem Afrikas, p. 209; among the Monbuttoos: 
Schweinfurth, Ztschr. f. Ethnol., V, p. 12; The Heart of Africa, II, pp. 
17-18; Casalis, Les Basoutos, p. 132. 



PRIMmVE ECONOMIC CONDITIONS. 39 

it behooves us, in our consideration of this minute eco- 
nomic separation, to guard against overlooking the unify- 
ing forces of working and caring for each other, and 
carefully as we must refrain from exaggerating the cen- 
trifugal forces here at work, it is nevertheless not to be 
denied that they are all traceable to one common origin — to 
the individual search for food practised through thousands of 
years by all these peoples. 

In this lies the justification of the method followed in 
this investigation, in which we have taken together peo- 
ples of very different stock and cultural stages, and con- 
sidered the economic phenomena separately. 

This procedure is, in political economy as in all social 
sciences, entirely justified; provided that, from the prodig- 
ious mass of disconnected facts that fill ethnology like a 
great lumber-room, we succeed in bringing a considerable 
number under a common denominator and rescuing them 
from the mystic interpretations of curiosity-hunters and 
mythologizing visionaries. For political economy in par- 
ticular this method offers the further and not inconsider- 
able advantage, that the toy mannikin in the form of a 
savage freely invented by the imagination of civilized man 
vanishes from the scene, and gives place to forms that are 
taken from life, although the observations from which they 
are drawn may leave much to be desired in point of ac- 
curacy. 

Our travellers have hitherto devoted little special atten- 
tion to the economy of primitive peoples. In the midst of 
their attention to dress, forms of worship, morals, religious 
beliefs, marriage customs, art, and technical skill, they 
have often overlooked what lay closest at hand, and in the 
gossipy records of ethnographic compilations the word 
" economy " has no more found a place than has the word 
" household " in the chronicles of the numerous investi- 



<^ - 



40 PRIMJTIt^E ECONOMIC CONDITIONS. 

gators into the constitution of the family. But just be- 
cause the observations we have utilized have been made 
for the most part only incidentally and not by trained 
economists, they possess a high measure of credibility. 
For they have on that account generally escaped the fate 
of being forced into some scheme categorically arranged 
in accordance with the conditions of our own civilization, 
and for that very reason unable to do justice to the differ- 
ently conditioned life of primitive peoples. 



CHAPTER II. 

THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 

The designation natural people seems a particularly- 
apt characterization of the lower races of men on their eco- 
nomic side. They stand in more immediate touch with 
nature than do we; they are more dependent upon her, 
are more directly susceptible to her powers, and succumb 
to them more easily. Civilized man lays by stores for the 
future; for the preservation and embellishment of his ex- 
istence he possesses a wealth of implements; in the event 
of failure of his crops the harvests of half a world stand at 
his disposal through our highly perfected means of trans- 
portation; he subdues the powers of nature and impresses 
them into his service. Our commerce places the labour 
of a thousand men at the command of every individual 
amongst us, and in every household watchful eyes guard the 
careful and economical consumption of the goods destined 
for our bodily subsistence. Primitive man, as a rule, gath- 
ers no stores; a bad harvest or other failure of the natural 
sources of his sustenance strikes him with its full weight; 
he knows no labour-saving implements, no system in dis- 
posing of his time, no ordered consumption; limited to 
his meagre natural powers, threatened on all sides by hos- 
tile forces, each day he has to struggle anew for his exist- 
ence, and often knows not whether the morrow will vouch- 
safe him the means to still his hunger. Yet he does not 
regard the future with anxiety, he is a child of the moment; 
no cares torment him; his mind is filled with a boundless 

naive egoism. With thoughts extending no further, he 

41 



42 THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 

instinctively follows his impulses, and in this regard also 
stands closer to nature than ourselves.^ 

The former rule was to classify primitive peoples accord- 
ing to the manner in which they procured their sustenance 
into hunters, fishers, pastorals, and agriculturalists. In 
this it was believed that each people must traverse these 
four stages of economic development in its progress 
towards civilization. The starting-point was the tacit as- 
sumption that primitive man began with animal food and 
only gradually passed over, under the stress of necessity, 
to a vegetable diet. The procuring of vegetable food 
moreover was considered the more difficult inasmuch as 
the picture of our European system of agriculture was ever 
in mind with its draught-animals and artificial apparatus 
of implements and tools. 

But this conception is erroneous, just as is the assump- 
tion from which it proceeds. Certainly all economic activ- 
ity begins with the procuring of food, which is wholly de- 
pendent upon the local distribution of the gifts of nature. 
From the beginning man was primarily dependent upon 
vegetable nourishment, and wherever tree-fruits, berries, 
and roots were to be gained, he first mac^e use of these. In 
case of need he turned to petty animals which could be 
consumed raw: shell-fish, worms, beetles, grasshoppers, 
ants, etc. Like the lower animal in continuous quest of 
food, he devoured at the moment what he found without 
providing for the future. 

If from this stage we seek the transition to the next, a 
little reflection tells us that it could not have been difficult 

* Comp. in general R. Vierkandt, Naturvolker u, KuUurvolker (Leip^ 
zig, 1896), pp. 260 ff. On the conception of natural peoples see further 
Panckow, Ztschr. d. Ges. f. Erdkunde zu Berlin, XXXI (1896), pp. 
158, 159. Anyone inclined to find fault with the indefiniteness of the 
definition should not overlook the fact that no single case has pre- 
sented itself raising the point whether a certain people was to be re- 
garded as primitive or not. 



THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 43 

to gain the practical knowledge that a buried bulb or nut 
furnishes a new plant,— certainly not more difficult than 
taming animals or inventing fish-hooks and bow and ar- 
row, which transition to the hunting stage required.^ As 
regards technical skill many hunting and nomadic peoples 
stand far above so-called agricultural peoples. Of late 
men have come to believe that nomads should rather be 
considered savage agriculturalists; and as a matter of fact 
it is highly improbable that a tribe of hunters should first 
have hit upon the taming of animals before they could pro- 
cure milk, eggs, and meat. Moreover, except in the ex- 
treme north, there is probably no fishing, hunting, or pas- 
toral people that does not draw a more or less consider- 
able portion of its sustenance from the vegetable kingdom. 
For this supply many of them have long been dependent 
upon trade with more highly developed neighboring peo- 
ples. They thus lack that economic independence which 
our study requires if it is to arrive at conclusions univer- 
sally applicable. 

' Now since the instances of hunting, fishing, and no- 
madic tribes, which are accepted as typical, are only to be 
found under special geographical and climatic conditions 
that hardly permit of a diflferent manner of obtaining 
sustenance (hunters and fishers in the farthest north, 
nomads in the steppes and desert places of the Old World), 
it may be advisable in our further consideration to leave 
them wholly aside and limit the field of our investigation 
to the intertropical districts £)f America, Africa, Australia, 
the Malay Archipelago, Melanesia, and Polynesia. This 
is still an enormous circuit, within which the diversity 
of natural conditions surrounding primitive man pro- 

' Comp. in general £. Hahn, Die Hausthiere «. ihre Besiehungen z, 
Wirihschaft d. Menschen (Leipzig, 1896) ; P. R. Bos, Jagdy Viehzuckt u. 
Ackerhau als KuUurstufen in Intern. Archiv f. Ethnographic, X (1897), 
pp. 187 ff. 



44 THE ECONOMIC LIFE OF PRlMlTiyE PEOPLE. 

duces many peculiarities in his material existence. The 
differences between the individual tribes in this regard are, 
however, not so great as, for instance, between the Esqui- 
maux and the Polynesian. At any rate, notwithstanding 
the great differences of the races in conditions of life and 
ways of living, they have still sufficient in common to oc- 
cupy our attention. In addition to this we have here 
the oldest regions inhabited by man, which, however, in 
spite, or perhaps on account, of the bounties of tropical 
nature, also appear to be those in which their development 
has been most slow. 

At all stages of his development the primitive man of 
these regions manifestly finds in vegetable diet the basis 
of his sustenance. This is evident from the simple 
fact that he has always found animal food much more dif- 
ficult to obtain. This is not contradicted by the circum- 
stance that at times we notice an eagerness for flesh break 
forth among many savage races which appalls us, since it 
does not shrink even from its own kind. The explanation 
in all probability is that the definite quantity of salt requi- 
site for the normal maintenance of the human body can- 
not be conveyed to it through purely vegetable diet, while 
it is quite possible with occasional raw-flesh food to live 
without salt. The same desire for salt is manifested even 
by the purely herbivorous among our domestic animals. 

The need of nourishment [we have seen] is the most 
urgent, and originally the sole, force impelling man to activ- 
ity, and causing him to wander about incessantly until it is 
satisfied. The species of division of labour between the 
two sexes, which this primitive search for food gives rise 
to, reaches its highest form when the wife procures the 
vegetable, the man the animal, portion of the food. And 
since, as a rule, the food gained is immediately devoured, 
and no one takes thought for the other as long as he him- 



THE ECONOMIC LIFE OF PRIhUTlVE PEOPLE. 45 

self is hungry, a difference in the nourishment of the two 
sexes arises that has perhaps contributed in an important 
degree to the differentiation of their bodily development. 

The division of labour of these primitive roaming hordes 
is continued at higher stages of development, and receives 
there such sharp expression that the rigidly limited spheres 
of activity of the man and the woman form almost a spe- 
cies of secondary sex-characteristics whose understanding 
gives us the key to the economic life of primitive peoples. 
In particular almost all their production of goods is dom- 
inated by it. 

Turning now to the latter, we should note by way of 
preface that by far the greater part of our primitive peo- 
ples, when they came under the view of Europeans, were 
acquainted with, and practised, agriculture. This is true, 
for instance, of all the negro races of Africa with but few 
exceptions, of the Malays, the Polynesians, and Mela- 
nesians, and of the primitive races of America, save those 
living at the extreme north and south of that hemisphere. 
It is a widely prevalent error, for which our youthful read- 
ing is responsible, that makes pure hunting races out of 
the North American Indians. All the tribes east of the 
Mississippi and south of the St. Lawrence were familiar 
with the cultivation of food-plants before the coming of 
Europeans; and, in the regions lying beyond, they at least 
gathered the grain of the water-rice {zizania aquatica) and 
ground meal from the berries of the manzanita shrub.* * 

The agriculture of primitive peoples is, however, pecu- 
liar.* In the first place it knows nothing of an implement 
that we think indispensable, namely, the plough. Wheel 
and wagon and draught-animals are likewise unknown. 
Furthermore, cattle-raising forms no integral part of their 

*Waitz, Anthropologie <f, Naturv'dlker, III, pp. 78 ff. 
* Comp. £. Hahn, as above, pp. 388 ff. 



46 THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 

agriculture. Fertilizing of the soil occurs at times, but it 
is extremely rare. More frequent are irrigation arrange- 
ments, especially for rice and taro plantations. As a rule, 
however, the cultivated land must be changed when its 
nutritive elements are exhausted; and the change is facili- 
tated by the absence of individual property in the soil, 
which belongs to the tribe or the village community as a 
whole. Lastly, [we may recall,] the preparing of the soil 
is almost exclusively woman's work. Only in the first 
clearing of a piece of land do the men assist. 

Of late years this system of culture has been designated 
the hack or hoe system, a short-handled hoe being its chief 
implement. With some tribes the primitive digging-stick 
still retains its place. At the basis of its plant production 
lie the tropical tuberous growths: manioc, yam, taro, sweet 
potato, pignut, and in addition bananas, various species of 
gourds (cucurbitacece), beans, and of the grains, rice, 
durra, and maize. Rice has its oldest home in South China 
probably, the durra in Africa, and maize, as is well known, 
in America. There belong, finally, to this system of agri- 
culture the tropical fruit-trees — sago-, date-, and cocoa- 
palms, the breadfruit-tree, and the like. 

On account of the imperfect nature and limited pro- 
ductivity of the implements, only small stretches of land 
can evfer be taken into cultivation under the hack system. 
It is closely related externally, and also in the manner in 
which it is carried on, to our garden culture. The fields 
are generally divided into beds, which are often hilled in 
an exemplary fashion and kept perfectly free of weeds. 
The whole is surrounded with a hedge to keep out wild 
animals; against the grain-eating birds, which are particu-' 
larly dangerous to the harvests of the tropics, the Malays 
set up very ingeniously constructed scarecrows; in most 
places in Africa special watch-towers are erected in the 



THE ECONOMIC LIFE OF PRIMITIVE PEOPLE, 47 

fields from which the young girls make noises to frighten 
away the animals. As a rule a definite crop rotation is ob- 
served. In the Congo basin, for instance, when the land 
is newly broken up it is first planted with beans, and when 
these are harvested, millet is sown; interspersed with the 
latter the sprouts of the manioc are often set out. The 
manioc does not yield full returns for from one and a half 
to two years, and occupies the land until the roots com- 
mence to become ligneous, and virgin soil must be taken 
up. In New Pomerania the rotation is yam roots first, 
then taro, and finally bananas, sugar-cane, and the like.® 

Travellers have often described the deep impression 
made upon them when, on coming out of the dreary 
primeval forest, they happened suddenly upon the well- 
tended fields of the natives. In the more thickly popu- 
lated parts of Africa these fields often stretch for many a 
mile, and the assiduous care of the negro women shines 
in all the brighter light when we consider the insecurity of 
life, the constant feuds and pillagings, in which no one 
knows whether he will in the end be able to harvest what 
he has sown. Livingstone gives somewhere a graphic 
description of the devastations wrought by the slave-hunts; 
the people were lying about slain, the dwellings were de- 
molished; in the fields, however, the grain was ripening, 
and there was none to harvest it. But as yet the life of 
these people is by no means firmly attached to the soil; 
seldom do their settlements remain for several generations 
on the same spot;^ their houses are fugitive structures of 

•Descriptions of hack agriculture in Angola, Congo district: Pogge, 
pp. 8, 9; Wissmann, Unter deutsch. Flagge, pp. 341 ff.; among the Mon- 
buttos: Schweinfurth, In the Heart of Africa, II, pp. 37—39; in 
Mindanao: Ztschr. f. Ethnol., XVII, pp. 19 ff.; in New Guinea: Finsch, 
Samoa fahrten, pp. 56 ff.; in New Pomerania: Parkinson, Im Bismarck- 
Archipelf pp. 118 ff.; in South America: Martins, Zur Ethnogr. Amer- 
ikas, pp. 84, 85, 489, 490. 

^Ratzel, I, p. 85; Panckow, pp. 167 ff. 



48 THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 

poles and grass; their other possessions may easily be 
carried away on their backs or quickly replaced, and on 
another spot a new village in a few days can be erected 
in which nothing from the old is lacking save the vermin. 

Hack agriculture is exactly suited to just such a life. 
It requires no fixed capital beyond the small hoe and, 
where corn is cultivated, perhaps a knife to cut the ears. 
The keeping of supplies is scarcely necessary, since in 
mai\y quarters the climate permits several harvests in the 
year. Where grain is grown, however, it is customary to 
store it in small granaries built on posts, or in pits in the 
earth, or in great earthen vessels. But even where thus 
stored it must soon be used if it is not to be destroyed 
through dampness, weevils, and termites. Livingstone 
thinks this explains why the negroes in case of abundant 
harvest brew so much beer.® 

Hack agriculture is still one of the most widely prevalent 
systems of husbandry. It is to be found throughout all 
Central Africa (iS** N. lat. to 22** S. lat.), in South and 
Central America, in the whole of the Australian islands, 
in great sections of Further India and the East Indian 
Archipelago. Everywhere, [as already noted], it appears 
to have been originally woman's work; and as such it is 
a great factor in advancing civilization. Obviously through 
hunting for roots, which she had practised from the earliest 
times, woman was led on to agriculture. Farinaceous bulbs 
and root-crops form accordingly the chief part of her plan- 
tations. In this manner she gained technical experiences 
which the man did not enjoy. Her labour soon yielded* 
the most important part of the requirements of life; and 
therewith was laid the foundation of a permanent family 
organization, in which the man undertook the offices of 
protection and the procuring of animal food. Only where 

^ Expedii, to Zambesi, p. 253. 



THE ECONOMIC UFE OF PRIMlTiyE PEOPLE. 49 

there are no large quantities of game does the man take 
part in the cultivation pf the soil, as, for instance, among 
the Malays. 

Let us now turn to the second source of sustenance, 
hunting and fishing. Through the imperfect nature of 
their weapons hunting has ever had among primitive peo- 
ples a strong resemblance to the method of the beast of 
prey stealing upon its victim. A larger animal can only be 
wounded by an arrow-shot or spear-thrust, not killed; and 
then it is the hunter's task to pursue the beast until it sinks 
down exhausted. As this species of hunt may, however, 
under certain circumstances, become very dangerous, the 
most varied ways of trapping have been invented — pits, 
barricades, and falling trees; or in attacking the animal 
directly the hunt is carried on by whole tribes or visage 
communities.' Under such circumstances communal own- 
ership of the hunting-grounds and the establishment of 
very detailed rules for the distribution of the booty among 
the participants and the owners of the ground have been 
early developed; but on such matters we cannot enter 
here.^® The essential thing for us to note is that the part of 
the duties pertaining to the providing of food necessitates 
a certain organization of work conformable to the princi- 
ple of labour in common — a circumstance that has cer- 
tainly been of the greatest importance for the birth of 
primitive political communities. 

The same is to be said of fishing, "^^ especially where the 
industry is followed along the seashore with boats and 
large nets, which can be produced and handled only with 
the help of many. The New Zealanders, for instance, wove 

* On the hunting methods of the Kaffirs comp. Fritsch, pp. 8i ff. ; of 
the South Americans, Martius, pp. 82, loi. 

"Some particulars in Post, Afr. Jurispr., II, pp. 162, 163; Lubbock, 
Origin of Civilisation^ p. 455. 

" As to fishing comp. in general Ratzel, I, pp. 234, 396, 506, 531. 



50 THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 

nets one thousand yards in length and it took hundreds of 
hands to use them. Innumerable are the modes of catch- 
ing fish which primitive peoples have invented; besides 
hook and net, arrows, spears, bow-nets, and methods of 
stunning the fish are resorted to. All our information on 
this subject indicates that fishing acquired a much more 
regular character among primitive peoples than hunting. 
On many islands of the South Sea, indeed, definite days of 
the week have once for all been set apart for the communal 
fishing; and the leaders of the fishing expeditions are also 
the leaders in war. Stream-fishing has been especially de- 
veloped by the primitive inhabitants of South America, 
among whom there are tribes that have been called fish- 
ing nomads because they wander from stream to stream. 
The same also occurs here and there in Africa. The actual 
labour of fishing seems always to fall to the men; it is only 
in some districts of Polynesia that the women take a lim- 
ited part in it. 

Because of the very perishable nature of meat, hunt- 
ing and fishing in tropical regions can, in the majority of 
cases, supplement the vegetable diet only occasionally. 
True, the drying and even the smoking of the fish and 
meat cut into strips was early learned and practised. This 
is the usage with the Polynesians as well as among the 
Malays and Americans, and even among the negroes and 
Australians. Yet the part of the food requirement which 
can be regularly met in this manner is so small that it is 
a rule among many tribes that only the more prominent 
persons may enjoy certain kinds of game. It is quite com- 
mon also for the use of certain kinds of meat to be for- 
bidden to the women. It is only small forest and coast 
tribes, who are able with their dried meat to carry on trade 
with agricultural neighbours, that find their support in 
hunting and fishing. 



y 



THE ECONOMIC LIFE OF PRlMlTiyE PEOPLE. 51 

It would accordingly be quite natural to assume that 
primitive peoples must early have hit upon the taming and 
raising of animals as a source of a regular food-supply. 
But we can speak of cattle-raising as a practice among the 
peoples of the tropical regions only in a very restricted 
sense. The hen alone, of our domestic animals, is to be 
found everywhere; besides it there is in Africa the goat, 
among the Malays and Polynesians the pig, and among 
the Americans the turkey, the musk-duck and the guinea- 
pig. Cattle are found oply among one section of the 
Malays and in one strip, more or less broad, of East 
Africa, which runs through almost the whole continent, 
from the Dukas and Bans on the Upper Nile to the 
Hottentots and Namaquas in the south. But most of 
these peoples do not use them as draught-animals; many 
of them do not even use their milk; many East African 
cattle-raisers never slaughter a beast ^except when they 
have captured it from another tribe.^^ Here and 
there in Equatorial Africa the ox serves as a riding and 
pack animal; but, generally speaking, the possession of 
cattle is for the negro peoples merely " a representation of 
wealth and the object of an almost extravagant venera- 
tion," — merely a matter of fancy. 

And this in general is the character of cattle-keeping 
among primitive peoples. An Indian village in the inte- 
rior of Brazil, [as we have remarked in our last chapter], 
resembles a great menagerie; even the art of dyeing the 
plumage of birds is known; but none of the many animals 
are raised because of the meat or for other economic pur- 
pose; the very eggs of the hens, which are kept in large 
numbers, are not eaten.*' To the Indian the lower animals 

" Schwcinfurth, I, pp. 59, 60; Livingstone, p. 553; Pogge, p. 23; 
Wissmann, Im Innem Afrikas, pp. 25, 127. 

"Ehrenreich, pp. 13, 14, 54; Martius, pp. 672 if.; K. v. d. Steinen, 
pp. 210, 379. Similarly among the Oceanians: Ratzel, I, p. 236. 



v^ 



52 THE ECONOMIC UFE OF PRIMITIVE PEOPLE. 

are beings closely related to man, and in which he delights; 
but, as is evident, this keeping of animals is much more 
closely allied to the hunt than to agriculture. Here we 
have to do with tamed, not with domestic, animals. With 
such a state of affairs the place of the pig in the domestic 
economy of the Oceanians has many cognate features; it is 
petted by the whole family; its young are not infrequently 
suckled by the women; and its flesh is eaten only on feast 
days and then by the more prominent people alone. The 
sole animal, [other than the hen], which is found among 
all primitive peoples is the dog; out it also is a pure luxury 
and is almost nowhere employed in the hunt; only a few 
tribes eat its flesh, and it has been claimed from ob- 
servation that these are always such as are devoted to 
cannibalism. 

On the whole, then, no importance can attach to cattle- 
raising in the production of the food-supplies of primitive 
peoples; in their husbandry it forms little more than an 
element of consumption. 

But the needs of these peoples are not confined to sus- 
tenance. Even the lowest among them paint, or in other 
ways decorate, their bodies, and make bows and arrows; 
the more advanced erect more or less substantial houses, 
plait and weave all kinds of stuffs, carve implements, make 
burnt earthen vessels; all prepare their food with fire, and 
with few exceptions know how to concoct intoxicating 
beverages. For all this labour of different kinds is neces- 
sary which we can characterize in a simple hianner as the 
transformation or working up of material, and which in the 
main embraces what we designate industry. Now what 
system did and does such work exhibit among primitive 
peoples? 

If we are to answer this question we must distinguish 
sharply the technical and the economic sides of industry. 



THE ECONOMIC UFE OF PfUMlTiyE PEOPLE. S3 

Technique in connection with the transformation of ma- 
terial is primarily dependent on natural conditions, and ac- 
cordingly develops among most primitive peoples only 
along special lines.*' Their implements are at first simple, 
natural objects, such as stones, animal bones, shells, sharp- 
ened pieces of wood, destined almost solely to increase the 
working power of the human members. Of implements 
consisting of more than one part, we may mention the 
hand-mill and crushing-mortar. The first is merely a sta- 
tionary and a movable stone with which the grains of c^orn 
are ground in the same manner as our artisans grind^aint 
in a mortar. The crushing-mortar is a hollowed tree-trunk 
with a wooden pestle. The simplest labour-saving me- 
chanical helps, such as wedge, lever, tongs, and screw, are 
unknown to them. Their boats are tree-trunks hollowed 
out with fire, or pieces of bark sewn together; the rudders 
are spoonlike pieces of wood with short handles represent- 
ing little more than a broadening out of the hand. The 
art of joining together pieces of wood or other hard mate- 
rial by pegs, nails, dovetailing, or glue they are ignorant 
of; for this purpose they use tough fibres or cords or even 
mere tendrils of climbing plants. Metal-working was un- 
known to the Australians, Melanesians, Polynesians, and 
the native inhabitants of America before the coming of the ' 
Europeans. On the other hand the negro peoples are uni- 
versally familiar with the procuring and working up of 
iron, and here and there of copper as well. A more ad- 
vanced technique as regards metals is found only among 
the Malays. But even in the iron-forging of the negroes 
all the technical awkwardness of these peoples can be per- 
ceived. Their smiths did not even hit upon the idea of 
making their own tools out of iron. Hammers and anvils 
are stones, and very often the tongs are only the rib of a 
palm-leaf. 

"On what follows comp. Arbeit u. Rhythmus, pp. lo ff 'f- 



54 THE ECONOMIC LIFE OF PRIMlTiyE PEOPLE, 

In spite of this technical backwardness many primitive 
peoples with their wretched tools produce wares of such 
quality and artistic taste as to arouse our highest admira- 
tion. This is possible only when the particular technical 
processes are applied in the simplest, and at the same time 
most comprehensive, manner. Preeminent are weaving, 
pottery, and wood-carving. What indeed do the tropical 
peoples not make out of the bast and fibrous material of 
their forests, of the tough grasses and rushes — from mats 
and clothing-stuffs of bark to water-tight baskets, dishes, 
and bottles? What is not made by the East Indians and 
Eastern Asiatics from bamboo — ^from the timbers of the 
house to water-vessels, blowing-tubes, and musical instru- 
ments? How highly developed is woodwork among the 
Papuans; and what patience and perseverance it demands! 
To weave a piece of stuff of raffia fibre in Madagascar often 
takes several months; and in South America the same time 
is required to finish a hammock. The polishing and pierc- 
ing of the milk-white pieces of quartz that the Uaupes 
of Brazil wear about their necks is frequently the work of 
two generations. 

This leads us directly to the industrial organization in the 
working up of material. For such labour there are, with 
few exceptions, no distinctly professional craftsmen. Each 
household has to meet all economic requirements of its 
members with its own labour; and this is accomplished 
by means of that peculiar division of duties between the 
two sexes, which we have come to know [in the preceding 
♦chapter]. Not only is it that a definite part of the provid- 
ing of food is assigned to either sex, but each looks 
after the preparing of such as is gained along with all at- 
tendant tasks. To the woman falls all that is connected 
with the procuring and preparing of the vegetable foods; 
to the man the making of weapons, and of implements for 
hunting, fishing, and cattle-raising, the working-up of ani- 



THE ECONOMIC LIFE OF PRJMlTiyE PEOPLE. 55 | 

mal bones and skins, and the building of canoes. As a rule i 

the man also looks after the roasting of the meats and the ' 

drying of fish, while the woman must attend to the labor- 
ious grinding of the com, which she has g^own, the brew- 
ing of beer, the shaping a^d burning of earthen pots for 
cooking, and in many instances the building of the huts as 
well. Besides these there are many species of the trans- 
forming of material, which are allotted now to one sex, 
now to the other. We may mention spinning, weaving, 
plaiting, the preparing of palm-wine and of bark stuffs. 
But, on the whole, this division of duties between the male 
and the female members of families is sharply drawn. In- 
deed it is continued even in consumption, for men and 
women never eat together; and, where polygamy exists, a 
separate hut must be provided for each wife.^** 

We cannot enter upon a more detailed discussion 
of this peculiarly evolved dualism in household economy 
among primitive peoples. It devolves upon us, however, 
to establish that the labour of the members of the house- 
hold, which is of such an individualistic character, cannot 
suffice for all tasks of their economic life. For under- 
takings that surpass the strength of the single household, 
assistance must therefore be obtained: either the help of the 
neighbours is solicited or all such labours are performed 
at one time by the whole village community. The latter 
is the rule in Africa, for instance, with the breaking of 
stretches of forest land for cultivation, the laying of barri- 
cades and pits for trapping wild animals, and elephant- 
hunting; in Polynesia, with the weaving of large fishing- 
nets, the building of large houses, the baking of bread- 
fruit in a common oven, and the like. Where clanship or 
slavery or polygamy exists, there is offered a means for 
multiplying the domestic working strength, and thus for 

** Comp. above, pp. 30 ff. -^ 



56 THE ECONOMIC LIFE OF PRJMlTiyE PEOPLE. 

accomplishing services of a higher order for which the 
individuars strength does not suffice. 

Within the separate tribes, accordingly, the working 
and refining of the raw products does not lead to the de- 
velopment of distinct trades, in that such work is carried 
on with uniform independence in each separate household 
From the reports of travellers, who judged from appear- 
ances, the existence of artisans among various primitive 
peoples has indeed been asserted. Thus on certain islands 
of the South Sea there are said to be professional carpen- 
ters, shoemakers, net-knitters, stone-borers, and wood- 
carvers. On closer examination of the particular cases 
these observations are open to doubt; to me the case of 
the native metal-worker seems alone to be proved. Among 
the negroes of Africa, as far as I can judge, only among 
the semi-civilized peoples of the Soudan are there the be- 
ginnings of a special industrial class. Beyond this any 
traces of a specialized industry supposed to have been dis- 
covered among primitive peoples are to be thus ex- 
plained: either individuals manifesting special aptitude 
for some manufacture came under the observation of 
travellers, or entire tribes excelled in a particular kind of 
household occupation, as we shall see directly. Trades 
formed only under European influence must naturally be 
disregarded here. 

But from tribe to tribe we find great differences in this 
industrial working-up of materials. It may even be safely 
claimed that almost every tribe displays some favourite 
form of industrial activity, in which its members surpass 
the other tribes. This is due to the varied distribution of 
natural products. If good potter's clay is to be found in 
the district or village of a particular tribe, the women of 
this tribe or village readily acquire special skill in pottery; 
where native iron ore is discovered, smiths will appear; 



THE ECONOMIC LIFE OP PRlMlTiyE PEOPLE. 57 

while along well-wooded seacoasts boat-building will flour- 
ish. Other tribes or localities excel in the preparation of 
salt from vegetable ashes, in the making of palm-wine or 
leather or skin garments; others again in the making of 
calabashes, baskets, mats, and woven materials. All these 
forms of skill, however, are aptitudes such as every man or 
every woman of the particular tribe or locality knows and 
also practises on occasion. When these individuals are 
designated by travellers as smiths, salt-makers, basket- 
makers, weavers, etc., that is to be taken in the same sense 
as when we speak of ploughmen, reapers, mowers, thresh- 
ers among our peasants, according to the work in which 
they are for the time engaged. We have here to do not 
with special trades claiming the individual's whole activity, 
but with arrangements forming essential parts of the econ- 
omy of each separate family. This naturally does not 
preclude single individuals from surpassing in skilfulness 
the other members of the tribe, just as there are among 
our peasant women particularly adept spinners, and among 
the farmers horse-breeders and bee-keepers who win in 
prize competitions. 

Travellers have often observed this tribal or local de^ 
velopment of industrial technique. " The native villages," 
relates a Belgian observer of the Lower Congo, " are often 
situated in groups. Their activities are based upon recip- 
rocality, and they are to a certain extent the complements 
of one another. Each group has its more or less strongly 
defined specialty. One carries on fishing, another pro- 
duces palm-wine; a third devotes itself to trade and is 
broker for the others, supplying the community with all 
products from outside; another has reserved to itself work 
in iron and copper, making weapons for war and hunting, 
various utensils, etc. None may, however, pass beyond 
the sphere of its own specialty without exposing itself to 



S8 THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 

the risk of being universally proscribed." From the 
Loango coast Bastian tells of a great number of similar 
centres for special products of domestic industry. Loango 
excels in mats and fishing-baskets, while the carving of 
elephants' tusks is specially followed in Chilungo. The so- 
called " Mafooka '* hats with raised patterns are drawn 
chiefly from the bordering country of Kakongo and May- 
yumbe. In Bakunya are made potter's wares which are in 
gjeat demand, in Basanza excellent swords, in Basundi 
especially beautiful ornamented copper rings, on the 
Zaire clever wood and tablet carvings, in Loango orna- 
mented cloths and intricately designed mats, in Mayumbe 
clothing of finely woven mat-work, in Kakongo embroid- 
ered hats and also burnt-clay pitchers, and among the Ba- 
yakas and Mantetjes stuffs of woven grass. 

Other similar accounts might be cited, not merely 
from Africa,*^ but also from the South Sea Islands and 
even from Central and South America,** We shall thus 
hardly err in assuming that in these tribal industries the 
controlling principle in the industrial development of prim- 
itive peoples has been discovered; that by them was 
furnished the means whereby the satisfaction of the needs 
of the individual and of whole groups was extended beyond 
their own immediate powers of production. For it may 
be taken for granted that an industrial product found only 
among those manufacturing it, especially if it attained to 
some importance in the simple life of these uncivilized peo- 
ples, would soon be coveted by the surrounding tribes. 

" H. Schuxtz has made a collection of them in his Afr, Gewerhe, pp. 
29-65. He has pursued further, though unfortunately not far enough, 
the subject of tribal industry. He has found such industry so exten- 
sively in evidence that we may assume the conditions of industrial 
production here portrayed to exist wherever travellers do not ex- 
pressly report to the contrary. 

" Comp. K. Sapper, Das nordl, MUteUAmerika (1897), pp. 299 £F., and 
the further examples given by us. 



THE ECONOMIC LIFE OF PRIMITIk'E PEOPLE. 59 

But the way from the coveting of an article to its enjoy- 
ment is a longer one in an economical organization, based 
upon acquisition directly by the individual himself, than 
we are inclined to assume in our own social life, which rests 
upon trade. 

In fact decidedly unclear conceptions are widely preva- 
lent as to the system of exchange of primitive peoples. We 
know that throughout Central Africa, from the Portuguese 
possessions in the west to the German in the east, there is a 
market-place every few miles at which the neighbouring 
tribes meet every fourth to sixth day to make mutual ex- 
changes. Of the Malays in Borneo we are told that each 
larger village possesses its weekly market. The first dis- 
coverers of the South Sea Islands give us reports of distant 
" trading trips " which the natives undertake from island 
to island in order to make mutual exchanges of their wares. 
In America certain products, the raw material for which is 
to be found only in a single locality — for example, arrow- 
points and stone hatchets made of certain kinds of stone 
— have been met with scattered throughout a great part of 
the continent.*'' Even among the aborigines of Australia 
there are instances of certain natural products, such as 
pitcher-plant leaves end ochre colour, which are found 
in but one place, and yet circulate through a great part 
of the country. In such phenomena we have a new and 
interesting proof of the civilizing power of trade; and 
in the primeval history of Europe itself this power has 
everywhere been assumed as operative when industrial 
products have been brought to light through excavations 
or otherwise far from their original place of production. 
Our prehistoric studies have woven together a whole 
spider's web of suppositions and have even brought us to 

^'Waitz, III, p. 75; on markets in South America, III, p. 380; on 
others in Mexico, IV, pp. 99 ff. 



^ 



6o THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 

speak of prehistoric " industrial districts." Our ethno- 
graphic literature speaks similarly of industrial localities 
for the manufacture of arms and the plaiting of mats in 
Borneo, for pottery at several points in New Guinea, for 
boat-building in several coast districts of the Duke of York 
Archipelago, for iron-working in negro countries, etc. 

In opposition to this it must be asserted positively that 
trade in the sense in which it is regarded by national econ- 
omy — ^that is, in the sense of the systematic purchase of 
wares with the object of a profitable re-sale as an organized 
vocation — can nowhere be discovered among primitive 
peoples. Where we meet native traders in Africa, it is a 
question either of intermediary activity prompted by Eu- 
ropean and Arabian merchants, or of occurrences peculiar 
to the semi-civilization of the Soudan. Otherwise the only 
exchange known to the natives everywhere is exchange 
from tribe to tribe. This is due to the unequal dis- 
tribution of the gifts of nature and to the varying develop- 
ment of industrial technique among the different tribes. 
As between the members of the same tribe, however, no 
regular exchange from one household establishment to 
another takes place. Nor can it arise, since that voca- 
tional division of the population is lacking which alone 
could give rise to an enduring interdependence of house- 
holds. 

One fancies the genesis of exchange to have been very 
easy because civilized man is accustomed to find all that 
he needs ready made at the market or store and to be able 
to obtain it for money. With primitive man, however, 
before he became acquainted with more highly devel- 
oped peoples, value and price were by no means current 
conceptions. The first discoverers of Australia found in- 
variably, both on the continent and on the neighbouring 
islands that the aborigines had no conception of ex- 



THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 6 1 

• 

change.^® The ornaments offered them had no power 
whatever to arouse their interest; gifts pressed upon 
them were found later on strewn about in the woods 
where they had been cast in neglect. Ehrenreich ^® and 
K. V. der-Steinen*^ had as late as 1887 the same experi- 
ence among the Indian tribes of Brazil. Yet there 
was from tribe to tribe a brisk trade in pots, stone hatchets, 
hammocks, cotton threads, necklaces of mussel-shells, and 
many other products. How was this possible in the ab- 
sence of barter and trade? 

 

The solution of this riddle is simple enough, and has 
now been confirmed by direct observation on the spot, 
while previously it could only be assumed. The transfer 
ensues by way of presents, and also, according to circum- 
stances, by way of robbery, spoils of war, tribute, fine, com' 
pensation, and winnings in gaming. As to sustenance, al- 
most a community of goods prevails between members of 
the same tribes. It is looked upon as theft if a herd of cat- 
tle is slaughtered and not shared with one's neighbour, or 
if one is eating and neglects to invite a passer-by. Anyone 
can enter a hut at will and demand food; and he is never 
refused. Whole communities, if a poor harvest befall, visit 
their neighbours and look to them for temporary support. 
For articles of use and implements there exists the uni- 
versal custom of loaning which really assumes the character 
of a duty; and there is no private ownership of the soil. 
Thus within the tribe where all households produce similar 
commodities and, in case of need, assist each other, and 
where surplus stores can only be utilized for consumption, 
there is no occasion for direct barter from establishment 
to establishment. Exceptions occur when purchasing a 

"Documentary proof in Ztschr. f. Sozial- u. Wirtschaftsgesch., IV, 
pp. 5 ff. (Sartorius v. Waltershausen). 
*• Beitrage s, Volkerkunde BrasiUens, p. 53. 
•• Unter d, Naturvolkem Central-BrasUiens (2d ed.), pp. 287 flF. 



6« THE ECONOMIC LIFE OF PRJMlTiyE PEOPLE, 

wife and making presents to the medicine-man, the singer, 
the dancer, and the minstrel, who are the only persons 
carrying on a species of separate occupations. 

From tribe to tribe there prevail rules of hospitality y^^ 
which recur with tolerable similarity among all primitive 
peoples. The stranger on arriving receives a present, 
which after a certain interval he reciprocates; and at his 
departure still another present is handed him.^^ On both 
sides wishes may be expressed with regard to these gifts. 
In this way it is possible to obtain things required or de- 
sired; and success is the more assured inasmuch as neither 
party is absolved from the obligations of hospitality until 
the other declares himself satisfied with the presents. 

That this custom of reciprocal gifts of hospitality per- 
mits rare products of a land or artistic creations of a tribe 
to circulate from people to people, and to cover just as 
long distances from their place of origin as to-day does 
trade, will perhaps become more apparent to us when we 
consider how legends and myths have in the same way 
been enabled to spread over half the world. It is almost 
inconceivable that this could have been so long overlooked 
when even in Homer the custom of gifts of hospitality is 
attested by so many examples. Telemachos brings home 
from Sparta as present from Menelaos a bowl of silver 
which the latter had himself received in Sidon as a gift of 
hospitality from King Phaidimos, and his father Odys- 
seus receives from the Phaiakes garments and linen and 
articles of gold as well as a whole collection of tripods and 
basins. All this he conceals on his arrival, as is well 
known, in the sacred grove of the nymphs in his native 
rocky island of Ithaca. Think of the poet's narration as 

^ On this point comp. K. Haberland, Die Gastfreundschaft auf nied, 
KuUurstufen: Ausland (1878), pp. 282 if. 

"Gift-making without recompense belongs only to a higher stage of 
civilization: A. M. Meyer, Ztschr. f. deutsch. Kulturgesch., V, pp. 18 ft. 



THE ECONOMIC LIFE OF PRIMITIf^E PEOPLE. 63 

an historical occurrence, and imagine what would have 
happened had Odysseus been recognised by the wooers at 
the right moment and slain; the presents of the Phaiakes 
would have rested well concealed in the grotto of the 
nymphs down to our own times, and would have been 
brought to light again by a modern archaeologist. Would 
he not have explained the whole treasure as the storehouse 
of a travelling merchant of the heroic age of Hellas, es- 
pecially as he could have appealed for support to the actual 
barter which occurs quite extensively in Homer? 

Among many primitive peoples peculiar customs have 
been preserved which clearly illustrate the transition from 
presents to exchange. Among the Dieris in Central Aus- 
tralia, for instance, a man or a woman undertakes for a 
present the task of procuring as reciprocal gift an object 
that another desires, or of hunting for him, or of per- 
forming some other service. The one thus bound is called 
yutschin, and until the fulfilment of the obligation wears a 
cord about his neck. As a rule the desired object is to be 
procured from a distance.** In New Zealand the natives 
on the Wanganui river make use of parrots, which they 
catch in great numbers, roast, and preserve in fat, in order 
to obtain dried fish from their fellow-countrymen in other 
parts of the island.*' Among the Indian tribes of Cen- 
tral Brazil trade is still an interchange of gifts of hospi- 
tality; and the Bakairis translate the Portuguese comprar, 
to buy, by a word signifying ' to sit down,' because the 
guest must be seated before he receives his present. In 
the countries of the Soudan the constant giving of pres- 
ents frequently becomes burdensome to the traveller 
*' since it is often only a concealed begging." " The gifts 
of hospitality that are received in the camp," remarks 

" A. W. Howitt in Journal of Anthrop. Inst., XX (1891), pp. 76 ff. 
" Shortland, Traditions and Superstitions of the New Zealanders (Lon- 
don, 1856), pp. 214, 215. 



64 THE ECONOMIC LIFE OF PRIMITIVE PEOPLE, 

Staudinger/^ *' are in accord with good custom and are 
often very welcome. But with every stop in a larger town 
things are frequently obtained from high and low which are 
ostensibly given as a mark of respect to the white man; 
in reality they arrive only because the donors expect a 
three- or four-fold response from the liberality of the Euro- 
pean. Indeed I am convinced that many a poor woman 
has herself first purchased the hen or duck that is to be 
presented in order to do a profitable piece of gift business 
with it." 

The Indians of British Guiana appear to stand at the in- 
termediate stage between gift-making and trading. Im 
Thurm reports of them: ^* 

" There exists among the tribes of this, as of probably every other 
similar district, a rough system of distribution of labour; and this 
serves not only its immediate purpose of supplying all the tribes with 
better-made articles than each could make for itself, but also brings 
the different tribes together and spreads among them ideas and news 
of general interest. . . . Each tribe has some manufacture peculiar 
to itself; and its members constantly visit the other tribes, often 
hostile, for the purpose of exchanging the products of their own 
labour for such as are produced only by the other tribes. The^e 
trading Indians are allowed to pass unmolested through the enemy's 
country. ... Of the tribes on the coast, the Warraus make far the 
best canoes, and supply these to the neighbouring tribes. They also 
make hammocks of a peculiar kind, which are not, however, much 
in request except among themselves. In the same way, far in the 
interior, the Wapianas build boats for all the tribes in that district. 
The Macusis have two special products which are in great demand 
amongst all tHe tribes. One is the ourali, used for poisoning arrows 
and the darts of blowpipes, the other is an abundance of cotton ham- 
mocks; for, though these are now often made by the Wapianas and 
True Caribs, the Macusis are the chief makers. The Arecunas grow, 
spin, and distribute most of the cotton which is used by the Macusis 
and others for hammocks and other articles. The Arecunas also sup- 
ply all blowpipes; for these are made of the stems of a palm which, 

"/m Her sen d. Haussalander (2d ed.), pp. 216, 217. Comp. Sachau, 
Reisen in Syrien u. Mesopotamien, p. 191; v. Hugel, Kaschmir, pp. 406, 

407. 
** Among ike Indians of Guiana (London, 1883), pp. 270-273. 



THE ECONOMIC UFE OF PRIMITIVE PEOPLE. 65 

growing only in and beyond the Veneznelan boundary of their terri- 
tory, are procured by the Arecunas, doubtless by exchange, from the 
Indians of the native district of that palm. The Tarumas and the 
Woyowais have a complete monopoly of the manufacture of the 
graters on which Indians of all the tribes grate their cassava. These 
two remote tribes are also the great breeders and trainers of hunting- 
dogs. . . • The True Caribs, again, are the most skilful potters; and 
though the Arawaks frequently, and the other Indians occasionally, 
make vessels for their own use, yet these are by no means as good as 
those which, whenever possible, they obtain from the Caribs. The 
Arawaks make fibr^ hammocks of a kind peculiar to them. . . . The 
Ackawoi alone, so far as I know, have no special product interchange- 
able for those of their neighbours. These Indians are especially 
dreaded and disliked by all the others; and it is possible that the want 
of intercourse thus occasioned between this tribe and the others forced 
the Ackawoi to produce for themselves all that they required. It is 
further possible that to this enforced self-dependence is due the miser- 
able condition of most of the Ackawoi. 

"To interchange their manufactures the Indians make long jour- 
neys. The Wapianas visit the countries of the Tarumas and the 
Woyowafs, carrying with them canoes, cotton hammocks, and now 
very frequently knives, beads, and other European goods; and, leav- 
ing their canoes and other merchandise, they walk back, carrying with 
them a supply of cassava-graters, and leading hunting-dogs, all which 
things they have received in exchange for the things which they took. 
The Macusis visit the Wapiana settlements to obtain graters and 
dogs, for which they give ourali-poison and cotton hammocks; and 
they again carry such of these gn'aters and dogs as they do not them- 
selves require, together with more of their own ourali and of their 
cotton hammocks, to other Indians — to the Arecunas, who give in 
return balls of cotton or blowpipes; or to the True Caribs, who pay 
in pottery." 

Once originated exchange long retains the marks of its 
descent in the rules that are attached to it and which are 
taken directly from the customs connected with gifts. This 
is manifested, in the first place, in the custom of payment 
in advance which dominates trade among primitive peo- 
ples.^^ The medicine-man does not stir his hand to help 

" Even European merchants trading in Africa must accommodate 
themselves to this custom 'by advancing to the black intermediaries 
whose services they call into requisition the price of the commodities 
that are to be supplied. Comp. Pogge, pp. 11, 140, 141; M. Buchner, 



66 THE ECONOMIC LIFE OF PRlMITiyE PEOPLE. 

the sick yntil he has received from the sick man's relatives 
his fee, which in this case closely resembles the present, 
and has openly announced his satisfaction. No purchase 
is complete until buyer and seller have before witnesses 
declared themselves satisfied with the objects received. 
Among many peoples a gift precedes or follows a deal ; *• 
the " good measures " of our village storekeepers, and 
"treating" are survivals of this custom. To decline 
without grounds an exchange that has been offered passes 
among the negroes as cm insult, just as the refusal of a 
gift among ourselves. The idea that services interchanged 
must be of equal value can hardly be made intelligible to 
primitive man. The boy who performs a bit of work ex- 
pects the same pay as the man, and the one who has 
assisted for one hour just as much as the one who has 
laboured a whok day; and as the greed on both sides 
knows no bounds, every trading transaction is preceded 
by long negotiations. Similar negotiations, however, are 
also the rule in the discharge of gifts of hospitality if the 
recipient does not find the donation in keeping with his 
dignity. 

As time passes exchange creates from tribe to tribe its 
own contrivances for facilitating matters. The most im- 
portant of these are markets and money. 

Markets are uniformly held among negroes, East Indi- 
ans, and Polynesians in open places, often in the midst of 
the primeval forests, on the tribal borders. They form neu- 
tral districts within which all tribal hostilities must cease; 
whoever violates the m^ket-peace exposes himself to the 
severest punishments. Each tribe brings to the market 

Kamerun, pp. 98, gg. Even the sacrifice to the deity seems to the peo- 
ples of this stage only payment in advance for an expected service: 
Heckewelder, Indian Nations, pp. 211 ff.; and comp. pp. 232, 236. 

" Schurtz, Entsteh. Gesch, d, Geldes, pp. 67, 68L Landor, In the For-' 
hidden Land (Tibet), I, p. 315; II, p. 78. 



THE ECONOMIC LIFE OF PRIMITIVE PEOPLE, 67 

whatever is peculiar to it: one honey, another palm- wine, 
a third dried meat, still another earthenware or mats or 
woven stuffs.^'' The object of the interchange is to ob- 
tain products that cannot be procured in one's own tribe 
at all, or at least ^cannot be produced so well and so artis- 
tically as in neighbouring tribes. This must- again lead 
each tribe to produce in greater quantities than it requires 
those products which are valued among the tribes not pro- 
ducing them, because in exchange for these it is easiest to 
obtain that which one does not possess one's self, but 
which others manufacture in surplus quantities. In each 
tribe, however, every household produces the current 
market commodity of exchange that enjoys this prefer- 
ence. Hence it follows, when it is a question of a product 
of house industry, such as earthenware or wares made of 
bark, that whole villages and tribal areas appear to travel- 
lers to be great industrial districts, although there are no 
specialized artisans, and although each household pro- 
duces everything that it requires with the exception of the 
few articles made only among other tribes which they have 
grown accustomed to and which exchange procures for 
them merely as supplements to houshold production. 

Such is the simple mechanism of the market among 
primitive peoples. Now with regard to money. How 
much has been written and imagined about the many 
species of money among primitive peoples,*® and yet how 

"Although many primitive peoples can be found ready to give 
everything for European wares that they have come to know and 
value, yet their regular exchange remains altogether one-sided and 
confined to a few articles. Many objects of daily use are not to be 
had from them at any price, especially objects of adornment. Comp. 
Finsch, Samoafahrten, pp. 108, 119, 236, 283 f., 315; Martius, cited 
above, pp. 89, 596; Ztschr. f. Ethnogr.. XVII, pp. 24, 62. 

"R, Andree, Ethnogr, ParalleUn u. Vergleiche, pp. 221 ff. O. Lenz, 
Ueber Geld bet d. Naturvolkem (Hamburg, 1895). F. Ilwof, Tausch^ 
kandel «. Geldsurrogate in alter «. neuer Zeit (Graz, 1882). H. Schurz, 



68 THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 

simple the explanation of their origin! The money of each 
tribe is that trading commodity which it does not itself 
produce, but which it regularly acquires from other tribes 
by way of exchange. For such article naturally becomes 
for it the universal medium of exchange for which it sur- 
renders its wares. It is its measure of value according to 
which it values its property, which could in no other way 
be made exchangeable. It is its wealth, for it cannot in- 
crease it at will. Fellow tribesmen soon come to employ 
it also in transferring values, for because of its scarcity it 
is equally welcome to all. Thus is explained what our trav- 
ellers have frequently observed, that in each tribe, often 
indeed from village to village, a different money is current, 
and that a species of mussel-shells or pearls or cotton stuff 
for which everything can be purchased to-day, is in the 
locality of the following evening's camp no longer ac- 
cepted by anyone. The consequence is that they must first 
purchase the current commodities of exchange before they 
can supply their own needs in the market. In this way, 
also, is to be explained the further fact, which has come 
under observation, that exchangeable commodities nat- 
urally scarce, such as salt, cauri shells, and bars of cop- 
per, or products of rare skill, such as brass wire, iron 
spades, and earthen cups, are taken as money by many 
tribes not possessing them; and above all is to be men- 
tioned the well-known circumstance of objects of foreign 
trade, such as European calicoes, g^ns, powder, knives, 
becoming general mediums of exchange. 

Certain varieties of money thus secure a more extensive 
area of circulation. They can even make their way into 
the internal trade of the tribal members through employ- 
ment as mediums of payment in the purchase of a bride, for 

Btitrdge s. Entstehungsgesch, d. Geldes: Deutsche Geogr, Blatter (B 
men), XX (1897), pp. 1-66. Intern. Archiv f. Ethnogn*., VI, p. 57. 



THE ECONOMIC LIFE OF PRIMITIVE PEOPLE, 69 

compensations, taxes, and the like; certain kinds of con- 
tracts are concluded in them. But there is no instance 
of a primitive people, in the absence of European influence, 
attaining to a currency or legal medium of payment for 
obligations of every kind and extent. It is rather the rule 
that various species of money remain in concurrent circu- 
lation; and very often certain obligations can be paid only 
in certain kinds. Changes in the variety of money are 
not infrequent; but on the other hand we sometimes find 
that a species will long survive the trade of the tribes from 
which it has gone forth, and will continue to serve in the 
inner transactions of a tribe, playing a singular, almost 
demoniacal, role, although, as regards their means of sus- 
tenance, the members of the tribe have nothing to buy and 
sell to one another. From an old interrupted tribal trade 
of this nature is to be explained the employment as money 
of old Chinese porcelain vessels among the Bagobos in 
Mindanao and the Dyaks in Borneo, the shells (dewarra) 
of the Melanesians, and the peculiar kinds of money of 
the Caroline Archipelago, for which special laws and ad- 
ministrative contrivances are necessary in order to keep 
this dead possession in circulation at all.^^ Otherwise the 
State does not interfere as a rule in these matters; and in 
the large territorial formations of Africa, such as the king- 
dom of Muata Yamwo, for instance, there are therefore 
diflferent currencies from tribe to tribe. But even where 
one kind of money gains a greater area of circulation, its 
value fluctuates widely at the various market-places; 
generally, however, it advances in proportion to the dis- 
tance from its source.'*^ 

** We cannot enter here more in detail into these matters, and would 
refer to the interesting descriptions of Kubary, Ethnog, Beitrage s, 
Kenntnis d. Carolinen-Archipels, pp. i fT., and Parkinson, pp. 79, 10 1 ff. 

"Thus Cecchi reports, Funf Jahre in Ostafrika, p. 271: "According 
to the higher or lower value of the salt bar in the markets of this 



70 THE ECONOMIC UFE OF PRIMITI^'E PEOPLE. 

Markets and money are intimately related so far as 
money in its character as a medium of exchange comes un- 
der consideration. But not every individual species of 
money that is met with among a primitive people has 
necessarily arisen from market trade. In its full develop- 
ment money is such an involved social phenomenon that 
it is natural to suppose that various influences associated 
with its past have been united in it. Thus, for instance, 
the origin of cattle-money seems to be bound up with the 
fact that, among the peoples referred to, the domestic ani- 
mals represented the wealth and the means of gathering 
wealth. That for the purchase of a bride and for similar 
ends many tribes do not receive the current money, but 
for such purposes prescribe certain other objects of worth, 
appears to point to the admissibility of the assumption that 
in the complete development of money, along with the 
main current, various subsidiary streams may have played 
a part.^^ 

From the standpoint of the total cultural progress of 

part of East Africa one could roughly estimate the distance from the 
place whence this money comes, and also judge of the more or less 
practicable nature of the routes over which the caravans transport it. 
Thus in the locality of its origin, for one thaler one receives among 
the Taltal, according to the statements of several travellers, several 
hundred salt bars. In Uorallu, the northern market of Schao, lying a 
distance of some two hundred miles from the country of the Taltal, 
its value fluctuates between fifteen and twenty for the thaler. In An- 
cober, eighty miles from Uorailu, the value sinks back to nine and 
nine and a half, and in Gera, two hundred and thirty miles beyond 
Ancober, one receives, according to circumstances, only six, five, 
four, or three salt bars per thaler." 

" Perhaps Karl Marx rightly expresses it when he tersely remarks: 
" The money-form attaches itself cither to the most important articles 
of exchange from outside, and these in fact are primitive and natural 
forms in which thccxchange- value of home products finds expression; 
or else it attaches itself to the object of utility that forms, like cattle» 
the chief portion of indigenous alienable wealth."— Ca^7a/ (Londoii» 
jfipi), p. 6i. 



THE ECONOMIC UFE OF PRIMITIVE PEOPLE. n 

mankind the most important result of this survey, how- 
ever, remains, that money as the favourite exchange com- 
modity furnished a medium that bound together men from 
tribe to tribe in regular peaceful trade, and prepared the 
way for a diflFerentiation of tribes in the matter of produc- 
tion. In the circumstance that all members of the same 
tribe or village preferably carried on, along with the earn- 
ing of their sustenance, other work of a definite type, lay 
the possibility of an advance in technical knowledge and 
dexterity. It was an international, or interlocal, division of . 
labour in miniature, which only much later was succeeded 
by division of labour from individual to individual within 
the nation, or the locality. Moreover the direct importance 
of the market for personal intercourse at this stage must 
not be undervalued, especially in lands where trading out- 
side the market is so unusual that even travellers wishing 
to buy something direct are regularly refused with the 
words " come to market." In this one is involuntarily re- 
minded of the prominent position that the market occu- 
pied in the social and political life of the peoples of classical 
antiquity. %>^ 

But it is always a very one-sided development, permitting 
only to individual tribes the organization of production 
and trade just described. In this way is to be explained 
that most extraordinary phenomenon that in the interior 
of continents where no difficulties in communication op- 
pose the passage of certain attainments in technical skill 
from tribe to tribe, it has been possible for peoples of 
very primitive economic stamp to remain unchanged by 
the side of others of higher development throughout 
thousands of years.^ One of the most remarkable exam- 
ples of this nature is offered in Central Africa by the pigmy 
race of the Batuas or Akkas, still standing at the stage of 
the lower nomads, which keeps strictiv within the zone of 



^ 



/ 



72 THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 

the primitive forest, but on definite days appears at the 
market-places of the surrounding negro tribes to exchange 
its chief economic product, dried meat of animals killed in 
the hunt, for bananas, ground-nuts, maize, and the like. In 
fact in some parts even a more primitive form of trading 
has been maintained between these pigmy people and their 
neighbours, in that at the period when thd fruit is ripe 
the Batuas break into the fields of the negroes, steal bana- 
nas, tubers, and corn, and leave behind an equivalent in 
meat.^* The fact that the Batuas are clever hunters ap- 
pears here to have caused the neighbouring tribes to neg- 
lect the production of meat through hunting and cattle- 
raising. On the other hand it is said that the pigmies do 
not even make their own weapons, but procure them iti 
trade from the Momsus and other tribes. 

Of this one-sided development another and much more 
wide-spread example is offered by the smiths, who not 
merely among many tribes of Africa but sporadically in 
Asia and in southeastern Europ>e form a hereditarily dis- 
tinct caste, whose members, whether regarded with bash- 
ful awe or contempt, can neither enter into a marital nor 

■*Casati, Zehn Jahre in Aeqttataria, I, p. 151. Schweinfurth, The 
Heart of Africa, II, pp. 83 ff. Dr. W. Junker's Travels in Africa, III, 
pp. 85, 86. Wissmann, Wolf, etc, Im Innem Afrikas, pp. 256, 258 ff. A 
similar report is given by W. Geiger (Ceylon), Tagebuchbldtter u, 
Reiseerinnerungen (Wiesbaden, 1897), of the Veddahs: "The method 
by which the Veddah is able to procure his arrow-points — which he 
'does not make himself — is interesting. He betakes himself under 
'Cover of night to the dwelling of a Singhalese smith, and places in 
front of it a leaf to which the desired shape is given. To this he adds 
a present of some kind, wild honey, the skin of an animal, or some- 
thing similar. During one of the following nights he returns and 
expects to find the object ordered finished. If he is satisfied, he will 
deposit another special gift. The smiths never refuse to execute the 
orders at once. If they do, they may be certain at the next oppor- 
tunity to be made the target for an arrow. Moreover their labour is 
abundantly rewarded by what the Veddah gives in return." 



THE ECONOMIC UFE OF PRlMlTiyE PEOPLE. 73 

Other social alliance with the rest of the people.** This 
strange phenomenon has hitherto been explained as a 
matter of remnants of subject tribes preserving to their 
conquerors the art of metal-working, which had otlier- 
wise perished, because the victorious race was ignorant of 
it. It is, however, also conceivable that a voluntary dis- 
persal of such tribes took place and that the very differ- 
ence of nationality, coupled with the carrying on of an 
esoteric art, placed them wherever they settled outside the 
community of the people. 

In individual instances the carrying on of such a tribal 
industry in this exclusive manner leads to the rise of what 
travellers usually designate now as industrial peoples, be- 
cause they do work for all their heighbours; now as trading 
peoples, because one meets them in all the markets of a 
more extensive district, and because they monopolize the 
trade in certain wares. We have an instance of the former, 
when the consumers resort to the district where a special 
tribal industry flourishes, in order to get the desired wares 
at the seat of manufacture; of the latter, when the pro- 
ducers bring to the tribes lacking them such wares as they 
produce beyond their own requirements. 

As an example of the first form of this evolution, the 
little tribe of the Osakas may be cited, which has its home 
in the valley of the' Ogowe to the east of the Lolo River. 
Lenz'* reports concerning it: "The Osakas are divided 
into five or six villages, each of which contains sixty to a 
hundred huts; compared with their numerically so im- 
portant neighbours, such as the Fans and the Oshebo- 
Adumas, they are thus destined to play an altogether 
passive role in the history of those countries. In spite of 
this, however, the Osakas appear to be not altogether in- 

" R. Andrec, pp. 153 ff. 

■• Mittheil. d. geogr. GcselL in Wien (1878), p. 476. 



74 THE ECONOMIC UFE OF PRIMITiyE PEOPLE, 

significant; for among them I found many individuals be- 
longing to the most widely different tribes, frequently 
from regions quite far distant. The Osakas are recognised 
as the best smiths, and all the surrounding tribes, — the 
Oshebo-Adumas, the Akelles, the Awanshis and even the 
Fans, — buy of them a great part of their implements for 
hunting and war, although the last-named tribe itself ex- 
cels at this handicraft. By the Oshebo-Adumas the iron 
wares of the Osakas are then brought down to the Okan- 
des and to the Apinshis and Okotas dwelling between the 
rapids of the Ogowe, these last tribes on their part being 
but little skilled in iron-work and devoting themselves ex- 
clusively to the slave-trade. From thete, through the me- 
dium of the Ining^s and Galloas, weapons of this kind find 
their way as far as the sea-coast." 

" The Oshebo-Adumas generally pay for these weap- 
ons with palm-oil and ground-nuts, while the Fans, who 
are the most expert huntsmen of all these various tribes, 
pve in exchange for the spears and swordlike knives dried 
and smoked meat, chiefly of the antelope, the' wild boar, 
the porcupine, the field rat and the monkey. In all the 
Osaka villages I saw a bustling life. As must always be 
the case where such widely different tribes meet together, 
quarrelings were extremely frequent there and often as- 
sumed great proportions." 

A typical example of the second form is offered by the 
Kiocos and the Kanjocas in the southern part of the 
Congo basin. Of the latter Wissmann reports: ^' " The 
Kanjoka country is particularly rich in iron, and there are 
some excellent smiths there. Salt also is produced, so that 
the Kanjokas, with the products of their country and their 
iron manufacture, undertake commercial expeditions to 
the south as far as the Lunda country." The Kiocos dwell 

"Afy Second Journey through Equat. Africa (London, 1891), p. 105. 



THE ECONOMIC LIFE OF PRIMITiyE PEOPLE. 75 

in the kingdom of Lund itself, dispersed among the Ka- 
lundas, but have their own chiefs who are tributary to the 
Muata Yamwo. The Kiocos are partial to placing their 
villages in the woodland, for they are preeminently excel- 
lent hunters, gather gum from their forests, and to obtain 
wax carry on a species of wild-bee keeping. They are also 
clever smiths, and not only make good hatchets, but can 
also repair old flintlocks and even fit them with new mounts 
and stocks. They clothe themselves in animal skins; the 
art of making vegetable cloths is little known to them. 
Their women plant chiefly manioc, maize, millet, ground- 
nuts, and beans. The products that the Kiocos obtain 
from the exploitation of their forests they exchange on the 
west coast for wares, chiefly powder, with which they then 
betake themselves into the far interior in order to buy 
ivory and slaves. The ivory obtained through trade they 
dispose of, while the slaves they procure they incorporate 
with their household. The Kiocos esteem slaves above all 
as property. They treat the slave women as they do their 
wives, and the men as members of the household, and part 
from them so very unwillingly that in the Kioco country 
it is quite exceptional for travellers to be offered slaves for 
sale. On their hunting voyages they have penetrated 
farthest towards the east; and there, before entering upon 
their journey homewards, they usually barter a part of 
their weapons for slaves. Then for the time being they 
arm themselves again with bow and arrow. They rightly 
enjoy the reputation of being as gxDod hunters as they are 
crafty and unscrupulous traders; and in a masterful man- 
ner they understand how to overreach and dispossess the 
better-natured and more indolent Kalundas.^^ 

This picture is often repeated in the negro countries. 

■• Pogge, pp. 45, 46, 47, and Wissmann, Im Innern Afrikas, pp. 59, 62. 
Comp. also Schurtz, Afr. Gew,, p. 50. 



76 THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 

One readily sees that it does not adapt itself to any of the 
usual categories of economic history. The Kiocos are no 
hunting people, no nomads, no agriculturalists, no indus- 
trial and trading nation; they are all these at once. They 
act as intermediaries for a part of the trade with the Euro- 
pean factories on the coast. At the same time they carry 
on some mediary traffic of their own in which they dis- 
play the peculiar aptitude of the negro for barter, but 
nevertheless gain most of their living directly from hunt- 
ing and agriculture. 

Both forms of development are met with in the two pot- 
tery islands of New Guinea, Bilibi and Chas. The manu- 
facture is in both places in the hands of the women. The 
natives of the islands round about, and even of the more 
distant ones, come to Chas to barter their products for the 
earthenware; in Bilibi the men take whole boatloads to sell 
along the coast. Every woman makes a special mark on 
the pottery she produces; but whether with one European 
observer we are to regard this as a trade-mark seems very 
doubtful.5* 

In order to leave untouched no important part of the 
economic life of primitive peoples, let us take a rapid 
glance at their commercial contrivances and public adminis- 
tration. Both are intimately connected. For commerce is 
essentially a public matter; there are no private commer- 
cial arrangements whatever among these peoples. Indeed 
one can claim frankly that at this stage trade scarcely dis- 
plays an economic character at all. 

In the first place as concerns commercial routes, there 
are overland trade routes only when they have been 
tramped by the foot of man; the only artificial structures 
to facilitate land trade are primitive bridges, often consist- 

**Comp. Finsch, pp. 82, 83, 281, 282; Semon, In the Austral, Bush, pp. 
317 ff. Similar pottery districts in Africa proven by Schurtz. p. 54. 



THE ECONOMIC UFE OF PRIMITIVE PEOPLE. 77 

ing merely of a single tree-trunk, or ferries at river fords, 
for the use of which the traveller has to pay a tax to the 
village chief. These dues as a rule open the door to heavy 
extortions.^^ On the other hand the natural waterways 
are ever3nvhere diligently used, and there is hardly a prim- 
itive people that has not been led through its situation 
by the sea or on a river to the use of some peculiar kind 
of craft. The enumeration and description of these means 
of transportation would fill a volume; from the dugout 
and skin canoe of the Indians to the artistically carved 
rowboats and sailboats of the South Sea Islanders, all 
types are represented. On the whole, however, the tech- 
nique of boat-building and navigation has remained unde- 
veloped among these peoples; none of their vessels de- 
serve the name of ship in the proper sense. Thus their 
importance is everywhere restricted to personal transporta- 
tion and fishing, while nowhere has the development 
reached a freight transportation of any extent. 

Curiously among primitive peoples that branch of com- 
mercial communication has enjoyed the fullest develop- 
ment which we would naturally associate only with the 
highest culture, namely, the communication of news. It 
forms indeed the sole kind of trade for which primitive 
peoples have created permanent organizations. We refer 
to the courier service and the contrivances for sending 
verbal messages. 

The despatching of couriers and embassies to neighbour- 
ing tribes in war and peace leads, even at a very low stage 
of culture, to the development of a complete system of 
symbolic signs and means of conveying intelligence.^® 
Thus among the rude tribes in the interior of Australia 

"Comp. Pogge, pp. 64, 70, 78, 95, 97, 115, 169; Wissmann, Unier 
dtutsch. Flagge, pp. 343, 361, 364, 394; and Second Journey, p. 71. 

" Comp. generally R. Andree, " Merkzeichen u. Knotenschrift " in 
his Ethnogr. Parallel, pp. 184 flF.; Waitz, IV, p. 89. 



78 THE ECuNOMiC LIFE OF PRIMITIVE PEOPLE. 

various kinds of body-painting, of head-dress and other 
conventional signs serve to apprise a neighbouring tribe of 
the occurrence of a death, of the holding of a feast, and of 
a threatening danger, or to summon the tribesmen to- 
gether for any purpose.** Among the aborigines of South 
America ingeniously knotted cords or leather strips 
(quippus), and among the North Americans the well- 
known wampum perform similar offices;*^ in Africa cou- 
rier-staffs with or without engraved signs are customary, 
and the same are found among the Malays and Poly- 
nesians. If need be, the couriers have to learn their mes- 
sage by heart and communicate it verbally.*^ In the negro 
kingdoms, where the administrative power of the ruler 
reaches only as far as he is able personally to intervene,** 
the couriers of the chiefs hold a very important position; 
for through them the sovereign chief is as if omnipresent, 
and new occurrences come to his knowledge with surpris- 
ing rapidity. But even for the communication of intelli- 
gence among members of the same tribe— for instance, in 
hunting and in war — a system of symbols exists which is 
often very ingeniously conceived, and which, as a rule, is 
hidden from the uninitiated. 

Not less remarkable are the telephonic contrivances rest- 
ing upon the ingenious employment of the drum, the 

Details in Journ. of Anthropol. Inst., XX, pp. 71 ff. 

Martius, pp. 98, 99, 694; Waitz, III, pp. 138 ff. On knot writing in 
West Africa: Bastian, Die Exp. n, d. Loango-Kiiste, I, p. 181. 

** Livingstone, p. 285. Comp. also the apt description by CasaJis, 
Les BassoutoSf pp. 234, 235: " Ces messagers sont generaJement doues 
d'une memoire prodigieuse, et Ton peut s'attendre i ce qu'ils trans- 
mettent textuellement les depeches orales, dont ils se chargent." 

''This applies also to the political conditions of semi-civilization. 
G. Rohlfs, Land u. Volk in Afrika, p. 163: **The Abyssinian is accus- 
tomed to obey only when his master is near. Once out of reach of his 
voice little does he trouble himself about orders. This is the case in 
all half-civilized countries. To this Turkey, Morocco, Egypt, and 
Bornoo bear witness." 



n 



THE ECONOMIC UFE OF PRIMITIVE PEOPLE. 79 

musical instrument in widest use among primitive peoples. 
In one sense they take the form of a developed signal- 
system, as among the East Indians** and the Melane- 
sians,*^ in another there is a real speaking of words by 
which detailed conversations can be carried on at great 
distances. The latter is very common in Africa.** As a 
rule only the chiefs and their relations are acquainted with 
this drum language; and the possession of the instrument 
used for this purpose is a mark of rank, like the crown and 
sceptre in civilized countries. Less extended is the em- 
ployment of fire-signs for summoning the tribe or com- 
municating news.** 

There is no public economy, in our sense of the word. 
True, where their power is to some extent established, the 
chiefs receive all kinds of dues in the form of shares, tra- 
ditionally fixed, in the products of the chase and of hus- 
bandry, fees for the use of bridges, ferries, market-place. In 
more extensive kingdoms the subordinate chieftains are 
bound to send tribute.** But all this is more or less mani- 
festly clothed in the form of gift, for which the chief has to 
bestow a return present even if this consist only in the 
entertainment that he bestows upon the bearer. Even 
with the market-fees, which are payable by the sellers to 
the owner of the market-place, in the Congo district a 

**Martius, p. 65. For a remarkable telephonic contrivance of the 
Catuquinani Indians, see Archiv f. Post u. Telegraphie (1899), PP> 
87, 8& 

'Parkinson, p. 127, comp. pp. 72, 121; Finsch, p. 68. Likewise in 
Africa: Schweinfnrth, I, pp. 64, 290, 291. 

^Described in greater detail by M. Buchner, Kamerun, pp. 37, 58; 
Wissmann, Jm Jnnem Afrikas, pp. 4, 22S, 232; Betz in MittheiL aus d, 
deutsch Schutzgebietejif XI (1898), pp. 1-86; Wissmann, Unter deutsch. 
Flagge, p. 215; Stanley, Through the Dark Continent, II, pp. 264, 279; 
Livingstone, p. 93. For a signal-whistling langruage in Timor, 
Jacobsen, Reise in d. InselweU d, Banda-Meers, p. 262. 

•Comp., for example, Petermanns MittheiL, XXI (1875), P- 381. 

' Details in Post, Afr, Jurisprudens, I, pp. 261 ff. 



So THE ECONOMIC LIFE OF PRIMITIk'E PEOPLE. 

return service is rendered in that the chief performs a 
dance in front, and to the delight, of those using the mar- 
ket. Of special interest to us are the presents that trav- 
ellers eil route have to pay to the village chiefs whose 
territories they traverse, since from such payments our 
customs duty has sprung. Not less important is it to notice 
that in the larger kingdoms the tribute of the subject 
tribes consists of those products which are peculiar to 
each tribe, and which are usually marketed by it. In the 
Lunda country, for instance, some districts bring ivory or 
skins, others salt or copper; from the northern parts come 
plaited goods of straw, and from the subordinate chiefs 
nearer the coast at times even powder and European cot- 
ton stuffs.*^ Not infrequently has this led such sovereign 
chiefs to carry on a trade in these products, which accumu- 
late in large quantities in their hands, or to claim a monop- 
oly in them. The saying that makes the kings the greatest 
merchants thus gains a deeper significance. 

In general the financial prerogatives of the chiefs are 
limited only by their natural strength; and the wealth of 
the subject is without the protection that the civilized 
State assures to it by law. The expeditions sent out by 
the negro kings to collect the tribute and taxes degen- 
erate only too often into robber raids. The claim of the 
kings to fines frequently reduces the administration of 
justice to an institution for extortion, and the system of 
gifts, which prevails in all relationships of a public char- 
acter, too rapidly passes into a veritable system of bribery. 

This must naturally react injuriously upon private in- 
dustry. In the condition of constant feud in which the 
smaller tribes live with their neighbours under the arbi- 

*^Pogge, pp. 226, 227. Comp. Wistimann, Im Innem Afrikas, pp. 
171, 172, 202, 249, 267, 286, 289, 308; Unter deutsch. Flagge, pp. 95, 332, 339. 
The same is true of the Marutse country north of the Zambesi: K 
Holub, Sieben Jahre in Sudafrika, II, pp. 173. 187, 253, 254, 257, 268, 271. 



THE ECONOMIC LIFE OF PRJMITIk'E PEOPLE. 8 1 

trary rule which in the interior usually accompanies the^ 
formation of larger states, most primitive peoples stand in 
peril of life and property. Through long habit this danger 
becomes endurable, yet economic advancement must as- 
suredly be retarded by it. The obligation to make pres- 
ents ever and everywhere, the custom of regarding food 
almost as free goods, leave but insufficient room for self- 
interest. An English writer makes the remark — from the 
standpoint of European life certainly not inaccurate — that 
this sharing-up, rendered necessary by custom, encourages 
the people in gluttony, since only that is safe which they 
have succeeded in stuffing into their bellies; it also pre- 
vents rational provision for the future, because it is diffi- 
cult to keep on hand supplies of any kind.^® Assuredly 
with some reason have the begging proclivities and the 
" tendency to steal," which is said to animate many prim- 
itive peoples in dealing with Europeans, been associated 
with the custom of gifts and the insufficient distinction of 
" mine and thine." ** The immoderate use of alcoholic 
drinks is likewise a consequence of their slight forethought 
for their own welfare. If, however, the attempt is made to 
appreciate all these things apart from the conditions of cul- 
ture in which they arise, one readily recognises that they 
lie " beyond the bounds of good and evil," and that what 
appears from the standpoint of the modern Englishman 
as vice has concealed within it the beautiful virtues of dis- 
interestedness, benevolence, and generosity. 

For many who to-day pose as the bearers of civilization 
to their black and brown fellow men primitive man is the 
quintessence of all economic vices: lazy, disorderly, care- 
less, prodigal, untrustworthy, avaricious, thievish, heart- 

^Undall, in Fritsch, p. 351; comp. p. 362. Waltz, II, p. 402; III, 
pw 80. 
*• Comp. Wait2, III, pp. 163 ff. 



82 THE ECONOMIC LIFE OF PRIMITIVE PEOPLE. 

less, and self-indulgent. It is true that primitive man lives 
only for the present, that he shuns all regular work, that 
he has not the conception of duty, nor of a vocation as a 
moral function in life. But it is not less true that with his 
wretched implements he accomplishes an amoimt of work 
that must excite our admiration, whether we contemplate 
with our own eyes the neat fruit-fields of the women or 
view in our museums the weapons and implements of the 
men, the products of infinite toil. Above all, his manner 
of working Assures to primitive man a measure of enjoy- 
ment in life and a perpetual cheerfulness which the Euro- 
pean, worried with work and oppressed with care, must 
envy him. 

If since their acquaintance with European civilization 
so many primitive peoples have retrograded and some 
even become extinct, the cause lies, according to the 
view of those best acquainted with the matter, chiefly in 
the disturbing influence which our industrial methods and 
technique have exerted upon them. We carried into their 
childlike existence the nervous unrest of our commercial 
life, the hurried hunt for gain, our destructive pleasures, 
our religious wrangles and animosities. Our perfected im- 
plements relieved them suddenly of an immense burden of 
labour. What they had accomplished with their stone 
hatchets in months they performed with the iron one in a 
few hours; and a few muskets replaced in effectiveness 
hundreds of bows and arrows. Therewith fell away the 
beneficent tension in which the old method of work had 
continuously kept the body and mind of primitive man, 
particularly as the character of his needs remained at the 
same low level. Under these conditions has he gone to 
ruin, just as the plant that thrives in the shade withers 
away when exposed to the glare of the noon-day sun. 



CHAPTER III. 

THE RISE OF NATIONAL ECONOMY. 

Everyone knows that the modern man's way of satisfy- 
ing his numerous wants is subject to continual change. 
Many arrangements and contrivances that we find neces- 
sary were unknown to our grandparents; and our grand- 
children will find inadequate much that perhaps only a 
short time ago aroused our admiration. 

All those arrangements, contrivances, and processes 
called forth to satisfy a people's wants constitute national 
economy. National economy falls again into numerous 
individual economies united together by trade and depend- 
ent upon one another in many ways; for each undertakes 
certain duties for all the others, and leaves certain duties 
to each of them. 

As the outcome of such a development, national econ- 
omy is a product of all past civilization; it is just as subject 
to change as every separate economy, whether private or 
public, and whether directly ministering to the wants of a 
larger or a smaller number of people. Furthermore, every 
phenomenon of national economy is a phenomenon in the 
evolution of civilization. In scientifically defining it and 
in explaining the laws of its development we must always 
bear in mind that its essential features and its dynamic 
laws are not absolute in character; or, in other words, that 
they do not hold good for all periods and states of civ- 
ilization. 

83 



84 THE RISE OF NATIONAL ECONOMY. 

The first task, then, which national economy presents 
to science is to determine and explain the facts. But it 
must not be content with a merely dynamic treatment of 
economic processes; it must also seek to deduce their 
origin. A full understanding of any given group of facts 
in the history of a civilized people requires that we know 
how the facts arose. We ^hall, therefore, not escape the 
task of investigating the phases of development through 
which the economic activity of civilized peoples passed be- 
fore it assumed the form of the national economy of to- 
day, and the modifications undergone by each separate 
economic phenomenon during the process. The material 
for this second part of the task can be drawn only from the 
economic history of the civilized peoples of Europe; for 
these alone present a line of development which historical 
investigation has adequately disclosed, and which has not 
been deflected in its course by violent disturbances from 
without; though, to be sure, this upward development has 
not always been without interruption or recoil. 

The first question for the political economist who seeks 
to understand the economic life of a people at a time long 
since past is this: Is this economy national economy; and 
are its phenomena substantially similar to those of our 
modern commercial world, or are the two essentially dif- 
ferent? An answer to this question can be had only if we 
do not disdain investigating the economic phenomena of 
the past by the same methods of analysis and deduction 
from intellectually isolated cases which have given such 
splendid results to the masters of the old " abstract " po- 
litical economy when applied to the economic life of the 
present. 

The modern " historical " school can hardly escape the 
reproach that, instead of penetrating into the life of earlier 
economic periods by investigations of this character, they 



THE RiSE OF NAT!0}^AL ECONOMY. . 85 

have almost unwittingly applied to past times the current 
classifications of modern national economy; or that they 
have kneaded away so long at conceptions of commercial 
life that these perforce appear applicable to all economic 
periods. In so doing they have without doubt greatly ob- 
structed the path to a scientific mastery of those historical 
phenomena. The material for economic history, which 
has been brought to light in such great quantities, has for 
this reason largely remained an unprofitable treasure still 
awaiting scientific utilization. 

Nowhere is this more plainly evident than in the manner 
in which they characterize the differences between the 
present economic methods of civilized nations and the eco- 
nomic life of past epochs, or of peoples low in the scale of 
civilization. This they do by setting up so-called stages 
of development, with generic designations made to embrace 
the whole course of economic evolution. 

The institution of such " economic stages " is from the 
point of method indispensable. It is indeed only in this 
way that economic theory can turn to account the 
results of the investigations of economic history. But 
these stages of development are not to be confounded 
with the time-periods of the historian. The historian 
must not forget to relate in any period everything 
that occurred in it, while for his stages the theorist 
need notice only the normal, simply ignoring the ac- 
cidental. In treating of the gradual transformation, fre- 
quently extending over centuries, which all economic phe- 
nomena and institutions undergo, his only object can be 
to comprehend the whole development in its chief phases, 
while the so-called transition-periods, in which all phenom- 
ena are in a state of flux, must, for the time, be dis- 
regarded. By this means alone is it possible to discover 



86 THE RISE OF NATIONAL ECONOMY. 

« 

the fundamental features, or, let us say it boldly, the laws 
of development. 

All early attempts of this class suffer from the defect of 
not reaching the essentials, and touching only the surface. 

The best known series of stages is that originated by 
Frederick List, based upon the chief direction taken by 
production. It distinguishes five successive periods which 
the peoples of the temperate zone are supposed to have 
passed through before they attained their present economic 
condition, namely: (i) the period of nomadic life; (2) the 
period of pastoral life; (3) the period of • agriculture ; (4) 
the period of combined agriculture and manufacture; and 
(5) the period of agriculture, manufacture, and commerce. 

Another series evolved by Bruno Hildebrand, which 
makes the condition of exchange the distinguishing char- 
acteristic, comes somewhat closer to the root of the mat- 
ter. It assumes three stages of development: period of 
barter; period of money; and period of credit. 

Both, however, take for granted that as far back as his- 
tory reaches, with the sole exception of the " primitive 
state," there has existed a national economy based upon 
exchange of goods, though at diflFerent periods the forms 
of production and exchange have varied. They have no 
doubt whatever that the fundamental features of economic 
life have always been essentially similar. Their sole aim 
is to show that the various public regulations of trade in 
former times found their justification in the changing char- 
acter of production or exchange, and that likewise in the 
present different conditions demand different regulations. 

The most recent coherent presentations of economic 
theory that have proceeded from the members of the his- 
torical school remain content with this conception, al- 
though in reality it stands upon a scarcely higher plane 



THE RISE OF NATIONAL ECONOMY. 87 

than the favourite historical creations of abstract English 
economics.^ This we will endeavour briefly to prove. 

The condition of society upon which Adam Smith and 
Rkardo founded the earlier theory is that of a commercial 
organization based upon division of labour;'^ or let us 
rather say simply, of national economy in the real sense of 
the term. It is that condition in which each individual 
does not produce the goods that he needs, but those which 
in his opinion others need, in order to obtain by way of 
trade the manifold things that he himself requires; or, in 
a word, the condition in which the cooperation of many 
or of all is necessary in order to provide for the individual. 
English political economy is thus in hs essence a theory of 
exchange. The phenomena and laws of the division of 
labour, of capital, of price, of wages, of rent, and of profits 
on capital, form its chief field of investigation. The whole 
theory of production and especially of consumption re- 
ceives very inadequate treatment. All attention is centred 
upon the circulation of goods, in which term their distri- 
bution is included. 

That there may once have existed a condition of society 
in which exchange was unknown does not occur to them; 
where their system makes such a view necessary they have 
recourse to the Robinson Crusoe fiction so much ridiculed 
by later writers. Usually, however, they deduce the most 
involved processes of exchange directly from the primitive 
state.^ Adam Smith supposes that* man is born with a 

* [Regarding the omission from special mention of Schmoller's ter- 
ritorial series: village, town, territory, and State, we may refer to 
Professor Schmoller's review of the first German edition and Profes- 
sor Bticher*s reply in Jhb. f. Gesetzgeb., etc., XVII and XVIII (1893- 
94). See also Schmoller, Grundriss d. Volkswirtschaftslehre, I (Leipzig, 
1900).— Ed.] 

' The same is true also of the Physiocrats. Comp. Turgot, RSHexions, 

SS2ff. 



88 THE RISE OF NATIONAL ECONOMY, 

natural instinct for trade, and considers the division of 
labour itself as but a result of it.* Ricardo in several places 
treats the hunter and the fisher of primitive times as if they 
were two capitalistic entrepreneurs. He represents them 
as paying wages and making profits; he discusses the rise 
and fall of the cost, and the price, of their products. Thu- 
netiy to mention also a prominent German of this school, in 
constructing his isolated State starts with the assumption 
of a commercial organization. Even the most distant 
region, which has not yet reached the agricultural stage, 
prosecutes its labours with the single end of selling its 
products in the metropolitan city. 

How widely such theoretical constructions vary from 
the actual economic conditions of primitive peoples must 
long ago have been patent to historical and ethnographical 
investigators had not they themselves been in the grasp 
of modern commercial ideas which they transferred to the 
past. A thorough-going study, which will sufficiently em- 
brace the conditions of life in the past, and not measure its 
phenomena by the standards of the present, must lead to 
this conclusion: National economy is thl^ product of a deveU 
opment extending over thousands of years, and is not older 
than the modern State; for long epochs before it emerged man 
lived and laboured without any system of trade or under forms 
of exchange of products and services that cannot be designated 
national economy. 

If we are to gain a survey of this whole development, it 
can only be from a standpoint that affords a direct view 
of the essential phenomena of national economy, and at 
the same time discloses the organizing element of the 
earlier economic periods. This standpoint is none other 
ttian the relation which exists between the production and 

V 

t 

4 

•Book I, Chap. 2. 



THE RISE OF NATIONAL ECONOMY, 89 

the consumption of goods ; or, to be more exact, the length 
of the route which the goods traverse in passing from pro- 
ducer to consumer. From this point of view we are able 
to divide the whole course of economic development, at 
least for the peoples of central and western Europe, where 
it may be historically traced with sufficient accuracy, into 
three stages: 

(i) The stage of independent domestic economy (produc- 
tion solely for one's own needs, absence of exchange), at 
which the goods are consumed where they are produced. 

(2) The stage of toivn economy (custom production, the 
stage of direct .exchange), at which the goods pass directly 
from the producer to the consumer. 

(3) The stage of national economy (wholesale production, 
the stage of the circulation of goods), at which the goods 
must ordinarily pass through many hands before they 
reach the consumer. - — -*• 

We will endeavour to define these three economic stages 
more precisely by seeking a true conception of the typical 
features of each without allowing ourselves to be misled 
by the casual appearance of transitional forms or particu- 
lar phenomena which, as relics of earlier or precursors of 
later conditions, project into any period, and whose exist- 
ence may perhaps be historically proved. In this way 
alone shall we be able to understand clearly the funda- 
mental distinctions between the three periods and the phe- 
nomena peculiar to each. 

The stage of independent domestic economy, as has already 
been pointed out, is characterized by restriction of the 
whole course of economic activity from production to con- 
sumption to the exclusive circle of the household (the 
family, the clan). The character and extent oi the produc- 
tion of every household are prescribed by the wants of its 
members as consumers. Every product passes through 



/ 



9© THE RISE OF NATIONAL ECONOMY. 

the whole process of its manufacture, from the procur- 
ing of the raw material to its final elaboration in the same 
domestic establishment, and reaches the consumer with- 
out any intermediary. Production and consumption are 
here inseparably interdependent: they form a single 
uninterrupted and indistinguishable process; and it is as 
impossible to differentiate them as to separate acquisitive 
and domestic activity from each other. The earnings of 
each communal group are one with the product of their 
labour, and this, again, one with the goods going to sat- 
isfy their wants, that is, with their consumption. 

Exchange was originally entirely unknown. Primitive 
man, far from possessing a natural instinct for trading, 
shows on the contrary an aversion to it. Exchange 
(tatischen) and deceive (tailschen) are in the older tongue 
one and the same word.* There is no universally recog- 
nised measure of valtie. Hence everyone must fear being 
duped in the bartering. Moreover, the product of labour 
is, as it were, a part of the person producing it. The man 
who transfers it to another alienates a part of his being 
and subjects himself to the evil powers. Far down into 
the Middle Ages exchange is protected by publicity, com- 
pletion before witnesses, and the use of symbolic forms. 

l\n autonomous economy of this kind is in the first place 
dependent upon the land under its control. Whether the 
chief as hunter or fisher appropriates the gifts voluntarily 
offered by nature, whether he wanders as a nomad with his 
herds, whether he cultivates the soil as well, or even sup- 
ports himself by agriculture alone, his daily labour and 
care will be shaped in every case by the bit of land that he 
has brought under cultivation. The greater his advance 

* tComp. also the early signification of our words barter, truck, etc. 
New Oxford Diet.— Ed.] 



THE RISE OF NATIONAL ECONOMY, 91 

in intelligence and technical skill, and the more method- 
ical and varied the satisfaction of his wants, so much the 
greater does this dependence become, until finally the soil 
brings into subjection the man who is bom to rule over it. 
This has been designated znllenage.^ We may here confine 
ourselves to proving that at this stage the man who has 
direct possession of the soil can alone maintain economic 
independence. He who is not in this position can eke out 
his existence only by becoming the servant of the land- 
owner, and, as such, bound to the soil. 

In the independent domestic economy the members of 
the household have not merely to gather from the soil its 
products, but they must also by their labour prodtrce all the 
necessary tools and implements, and, finally, work up and 
transform the new products and make them fit for use. All 
this leads to a diversity of employments, and, because of 
the primitive nature of the tools, demands a varied dex- 
terity and intelligence of which modern civilized man can 
scarcely form a proper conception.^ The extent of the tasks 

* Verdinglichung. 

' We must turn to descriptions of early peasant life in remote parts 
of Europe in order to gain a conception of such conditions. Comp. 
one example in H. F. Tiebe, Lief- u, EstMands Ehrenrettung (Halle, 
1804), p. icx>. Similar instances are met with still among the Coreans. 
Thus we read in M. A. Pogio, Korea (Vienna and Leipzig, 1895), p. 
222: "Throughout Corea the real necessaries of life have been pro- 
duced within the household from time immemorial. The wife and 
daughters spin not only hemp but silk. For the latter a silk-bee is 
usual in many houses. The head of the family must be ready for all 
tasks, and on occasion play the painter, stone-mason, or joiner. The 
production of spirits, vegetable fats, and colours, and the manufacture 
of straw mats, hats, baskets, wooden shoes, and field implements be- 
longs to domestic work. In a word, every one labours for himself and 
his own requirements. Thanks to these conditions the Corean is a 
Jack of all trades who undertakes work only for the things that are in- 
dispensable, and accordingly never becomes skilled in any special de- 
partment 



92 THE RISE OF NATIONAL ECONOMY. 

falling to the various members of this autonomous house- 
hold community can be lessened only by division of labour 
and cooperation among themselves according to age and 
sex, or according to the strength and natural aptitudes of 
the individual. It is to this circumstance that we must 
ascribe that sharp division of domestic production accord- 
ing to sex, which we find universal among primitive peo- 
ples. On the other hand, owing to the unproductiveness 
of early methods of work the simultaneous cooperation of 
many individuals was in numerous instances necessary to 
the accomplishment of certain economic ends. Labour in 
common still plays, therefore, at this stage, a more im- 
portant role than division of labour. 

To neither, however, would the family have been able 
to give much scope had it been organized like our modern 
family, that is, limited to father and mother with children 
and possibly servants. It would also have had very little 
stability or capacity for development if each individual in 
the family had been free to lead the independent existence 
of the present day. 

Significant is it then that when the present civilized na- 
tions of Europe appear on the horizon of history, the 
tribal constitution prevails among them.^ The tribes 
(families, gentes, clans, house communities) are moder- 
ately large groups consisting of several generations of 
blood-relations, which, at first organized according to ma- 
ternal and later according to paternal succession, have 
common ownership of the soil, maintain a common house- 
hold, and constitute a union for mutual protection. Every 
tribe is thus composed of several smaller groups of rela- 

* Comp. on this point Fustel de Coulanges, La citi antique (Paris, 
1864); Emile de Laveleye, De la ProprUti (4th edition, Paris, 1891); 
E. Grosse, Die Formen d. Familie u. d. Formen d. Wirthschaft (Leipzig, 
1896), especially Chap. VIII. 



THE RISE OF NATIONAL ECONOMY. 93 

tives, each of which is formed of a man and wife with their 
children. Anyone living outside this tribe is an out- 
law; he has no legal or economic existence, no help in time 
of need, no avenger if he is slain, no funeral escort when 
he passes to his last rest.^ 

All the peoples in question, when they took up fixed 
abodes, were acquainted with the use of the plough. Their 
settlement came about usually by the establishment of 
large common dwelling-houses, farms, and villages by the 
members of a tribe. Once in secure possession of the land 
the sense of community soon weakened. Smaller patri- 
archal households with a limited number of members, such 
as are represented at the present day by the zadrugas of 
the south Slavs, and by the great family of the Russians, 
Caucasians, and Hindoos, separated from the larger unit. 
But for centuries the village house-communities continued 
to own the soil in common, and jointly tilled it probably 
for some time longer, while each household enjoyed the 
products apart. 

In large family groups of this kind, community and divi- 
sion of labour may be carried out to a considerable extent. 
Men and women, mothers and children, fathers and grand- 
fathers — to each group is allotted its particular part in pro- 
duction and domestic work, and wherever special individ- 
ual skill displays itself, it finds scope and also a limit, in 

*Comp. M. Bttchner, Kamerun, p. 188: "It is a fundamental point 
in the legal conceptions of the negroes, that not the man himsfelf but 
the community, the family, the whole body of relatives is the individ- 
ual before the law. Within the community rights and duties are 
transferable to an almost unlimited extent. A debtor, a criminal, can 
be punished in the members of his community, and the liability of 
the community for the crime of one born a member of it does not 
lapse even with emigration or separation from it. Even the death- 
penalty can be executed upon one other than the guilty." The same 
thing is found among the South Sea Islanders. See Parkinson, Im 
Bismarck'Archipel, pp. 80-1. 



94 THE RISE OF NATIONAL ECONOMY. 

working for its own tribe. The feelings of brotherhood, 
of filial obedience, of respect for age, of loyalty and defer- 
ence reach their most beautiful development in such a 
community. Just as the tribe pays a debt or weregild for 
the individual or avenges a wrong done him, so on the 
other hand does the individual devote his whole life to the 
tribe and on its behalf subdue every impulse to independ- 
ent action. 

And even when the strength of these feelings declines, 
the modern separate family with its independent organiza- 
tion does not immediately spring into existence. For its 
appearance would inevitably have resulted in a diminished 
capacity for work, an abandonment of the autonomous life 
of the household, and perhaps a relapse into barbarism. 
Two ways there were of avoiding this. 

One was as follows: for such tasks as surpassed the pow- 
ers of the now diminished family, the original large tribal 
unions were continued as local organizations. These 
formed partial communities on the basis of common prop- 
erty and common usufruct of the same; but, when occa- 
sion demanded, they could also undertake duties which, if 
left to the care of each individual household, would have 
demanded an unprofitable expenditure of energy, as, for 
example, guarding the fields and tending cattle. There 
were also tasks which, though not of equal concern to 
each separate household of the local group, were never- 
theless too difficult for the individual. A house or a ship 
was to be built, a forest clearing made, a stream diverted, 
hunting or fishing engaged in at a distance; or perhaps 
the season of the year made some unusual work neces- 
sary for this or that house. In all such cases bidden- 
labour assisted; • that is, among neighbours there sprang 

* Cotnp. Arbeit «. Rhythmus (2d ed.). pp. 198 ff. [and Gl VII, be- 
low. — Ed.] 



THE RISE OF NATIONAL ECONOMY. 95 

up, on invitation of the head of the family, temporary labour 
communities which disappeared again on the completion of 
their work. Many institutions of this kind underwent subse- 
quent transformation, others perpetuated themselves. We 
would recall the labour communities of the Slavic tribes, 
the artel of the Russians, the tscheta or druzina of the Bul- 
garians, the moba of the Serbs, the voluntary assistance 
rendered by our peasants to each other in house-raising, 
sheep-shearing, flax-pulling, etc. 

Whatever the extent of such contrivances, the part they 
can play in the supplying of needs is comparatively un- 
important, and just as little prejudices the economic auton- 
omy of the individual household as the home production 
subsisting among our agrarian landlords to-day affects the 
supremacy of commerce. These temporary labour com- 
munities, moreover, are not business enterprises, but only 
expedients for satisfying immediate wants. Assistance is 
rendered now to one, now to another of the participants; 
or the product of the joint labour is distributed to the sep- 
arate families for their consumption. A definite case of 
bargain and sale will be sought for in vain, even where, as 
in the village community of India, we have a number of 
professional labourers performing communal functions 
similar to those of our village shepherds. They work for 
all and are in return maintained by all. 

The second method of avoiding the disadvantages aris- 
ing from the dissolution of the tribal communities con- 
sisted in the artificial extension, or numerical maintenance 
of the family circle. This was done by the adoption and 
incorporation of foreign (non-consanguinous) elements. 
Thus arose slavery and serfdom. 

We may leave undecided the question whether the en- 
slavement and setting to work of a captured enemy were 
more the cause or the result of the dissolution of the early 



gO THE RISE OF NATIONAL ECONOMY. 

tribal community. It is certain that a means was thereby 
found of maintaining intact the independent household 
economy with its accustomed division of labour, and at the 
same time of making progress towards an increase in the 
number and variety of wants. For now the more numer- 
ous the slaves or villeins belonging to the household, the 
more completely could its labour be united or divided. In 
agriculture larger areas could be cultivated. Particular 
technical employments, such as grinding com, baking, 
spinning, weaving, making implements, or tending cattle, 
could be assigned to |>articular slaves for their whole life; 
they could be specially trained for this service. The 
more prominent the family, the more wealthy the lord, or 
the more extensive his husbandry, all the more possible 
was it to develop in variety and extent the technical skill 
employed in the procuring and working up of materials. 
The economic life of the Greeks, the Carthaginians, and 
the Romans was of this character.'' Rodbertus, who notii:ed 

* For students of political economy it need scarcely be observed 
that in what follows the object is not to furnish a compendium of the 
econc^mic history of ancient times, but, as the context shows, merely 
an outline of the most highly develpped domestic economy as it 
presents itself in the system of slave labour among the ancients. In 
my work on the insurrections of the unfree labourers between 143 and 
129 B.C. (Die Aufstdnde d. unfreien Arbeiter, 143-12^ v. Chr., Fr.-a.-M., 
1874), I have shown that before the rise of slave-work on a large 
scale the economic life of antiquity furnished considerable scope for 
free labour, the formation of separate trades, and the exchange of 
goods. What progress had been made in the development of an in- 
dependent industry, I have set forth in the article " Industry " 
{Gewerhe) in the Handworterbuch der Staatsw., Ill, pp. 926-7, 92^ 
931; and in my articles on the Edict of Diocletian on tax prices 
(Ztschr. f. d, ges, Staatsw. ^ 1894, pp. 200-1) I have endeavoured to fix 
the position filled by trade in the system of independent domestic 
economy at the time of the empire in Rome. Reference may also be 
made, for an outline picture of the times, to the interesting address 
of M. Weber on Die sosialen Grunde d. Unier gangs d, antiken Cultur, Die 
Wahrheit, VI, No. 3. 



THE RISE OF NATIONAL ECONOMY. 97 

this a generation ago, designates it oikos husbandry, be- 
cause the oiKoiy the house, represents the unit of the eco- 
nomic system. The oiKoi is not merely the dwelling-place, 
but also the body of people carrying on their husbandry in 
common. Those belonging to them are the oiscitat, a 
word which, in its historic usage, it is significant to note, 
is confined to the household slaves upon whom the whole 
burden of the work of the house at that time rested. A 
similar meaning is attached to the Roman familia, the 
whole body of famuli^ house-slaves, servants. The pater- 
familias is the slave-master into whose hands flows the 
whole revenue of the estate; in the patria potestas the two 
conceptions of the power of the lord as husband and father 
and as slave-owner have been blended. A member of the 
household labours not for himself, but only for the pater- 
familias, who wields the same power of life and death 
over all. 

Jn the supreme power of the Roman paterfamilias, ex- 
tending as it did equally over all members of the house- 
hold, whether blood-relatives or not, the independent do- 
mestic economy was much more closely integrated and 
rendered capable of much greater productivity than the 
matriarchal or even the earlier patriarchal tribe, which 
consisted solely of blood-relatives. The individual as a 
separate entity has entirely disappeared; the State and the 
law recognise only family communities, groups of per- 
sons, and thus regulate the relations of family to family, 
not of individual to individual. As to what happens within 
the household they do not trouble themselves. 

In the economic autonomy of the slave-owning family 
lies the explanation of all the social and a great part of the 
political history of Rome. There are no separate classes 
of producers, as such, no farmers, no artisans. There 
are only large and small proprietors, rich and poor. If the 



98 THE RISE OF NATIONAL ECONOMY. 

rich man wrests from the poor possession of the soil, he 
makes him a proletarian. The landless freeman is 
practically incapable of making a living. For there is no 
business capital to provide wages for the purchase of 
labour; there is no industry outside the exclusive circle 
of the household. The artificers of the early records are 
not freemen engaged in industry, but artisan slaves who 
receive from the hands of the agricultural and pastoral 
slaves the corn, wool, or wood which are to be transformed 
into bread, clothing, or implements. " Do not imagine 
that he buys anything," we read in Petronius of the rich 
novus homo, " everything is produced at his own house." * 
Hence that colossal development of latifundia, and, con- 
centrated in the hands of individual proprietors, those end- 
less companies of slaves amongst whom the subdivision of 
labour was so multiplex that their productions and ser- 
vices were capable of satisfying the most pampered taste. 
The Dutchman, T. Popma, who in the seventeenth ^n- 
tury wrote an able book on the occupations of the Roman 
slaves, enumerates one hundred and forty-six different 
designations for the functions of these slave labourers oi 



'Sat. 38: ** Nee est quod putes ilium quicquam emere; omnia domi 
nascuntur." E. Meyer translates that, "everything is grown on his 
own land"! Now the satirist specifies wool, wax (?), pepper, sheep, 
honey, mushrooms, mules, and cushions with covers of purple or scar- 
let. Do all of these things grow from the soil? Compare also Petro- 
nius, ch. 48, 52, and 53 : " nam et comoedos emeram," etc. That this 
is all greatly exaggerated it is unnecessary to remind anyone who has 
really read Petronius. Ch. 50 speaks of the purchase of Corinthian 
jars; ch. 70 of knives made of Noric iron bought in Rome; ch. 76 
of the shops of Triraalchio, who himself gives as his motto the words 
bene emo, bene vendo. But for a satirtist to venture such an exaggera- 
tion as Petronius in ch. 38 would have been impossible if Roman eco- 
nomic life had been similar to that of to-day. A modem satirist in 
a similar case would have made his boaster give the values of his 
horses, wines, cigars, his stocks, etc. 



THE RISE OF NATIONAL ECONOMY. 99 

the wealthy Roman households.* This number might to- 
day be considerably increased from inscriptions. One 
must go minutely into the details of this refined subdivi- 
sion of labour in order to understand the extent and pro- 
ductive power of those gigantic household establishments 
that placed at the free disposal of the owner goods and 
services such as to-day can be supplied only by the numer- 
ous business establishments of a metropolitan city in con- 
junction with the institutions of municipality and State. 
At the same time this extensive property in human beings 
afforded a means for the amassing of fortunes equalled 
only by the gigantic possessions of modem millionaires. 

The whole body of slaves in the house of a wealthy 
Roman was divided into two main groups, the familia 
rustica and the familia urbana. The familia rustica en- 
gages in the work of production. On every large country 
estate there are a manager and an assistant manager with 
a staff of overseers and taskmasters who in turn have un- 
der them a considerable company of field-labourers and 
vine-dressers, shepherds and tenders of cattle, kitchen and 
house servants, women spinners, male and female weavers, 
fullers, tailors, carpenters, joiners, smiths, workers in metal 
and in the occupations connected with agriculture. On 
the larger estates each group of labourers is again divided 
into bands of ten each (decuricf) in charge of a leader or 
driver (decurio, monitor)}^ 

The familia urbana is divided into the administrative 
staff, and the staff for the service of master and mistress 
within and without the house. First comes the superin- 
tendent of the revenue with his treasurer, bookkeepers, 

•Titi Popmae Phrysii de operis servorum liber. Editio novissima. 
Amstelodami 1672. 

"Comp. the graphic account of work on a Roman estate during 
the empire, by M- Weber, Die Wahrheit, VI, pp. 65, 66. 



loo THE RISE OF NATIONAL ECONOMY. 

supervisors of rents, buyers, etc. If the proprietor takes 
over public leases or engages in the shipping trade, he 
keeps for that purpose a special staff of slave officials and 
labourers. Attached to the internal service of the house 
are house-administrator, porters, attendants in rooms and 
halls, guardians of the furniture, the plate, and the robes; 
the commissariat is in charge of the steward, the cellar- 
master, and the superintendent of supplies; the kitchen 
swarms with a great company of cooks, stokers, bakers of 
bread, cakes, and pastry; special table-setters, carvers, 
tasters, and butlers serve at the table, while a company of 
beautiful boys, dancing-girls, dwarfs and jesters amuse the 
guests. To the personal service of the proprietor are as- 
signed a master of ceremonies for introducing visitors, 
various valets, bath attendants, anointers, rubbers, sur- 
geons, physicians for almost every part of the body, 
barbers, readers, private secretaries, etc. For service in 
the household a savant or philosopher is kept, also archi- 
tects, painters, sculptors, and musicians ; in the library are 
copyists, parchment-polishers, and bookbinders, who un- 
der the direction of the librarian make books in the private 
manufactory of the house. Even slave letter-writers and 
stenographers must not be wanting in a wealthy house.^® 
When the master appears in public he is preceded by a 
large body of slaves (anteambulones), while others follow 
him (pedisequi); the nomenclator tells him the name of those 
whom he meets and who are to be greeted; special dis- 
tributores and tesserarii scatter bribes among the people 
and instruct them how to vote. These are the camelots of 
ancient Rome; and, what gives them special value, they 
are the property of the distinguished aspirant employing 
them. This system for exerting political influence is sup- 



19 



Comp. ch. 6. 



THE RISE OF NATIONAL ECONOMY. loi 

plemented by the institution of plays, chariot-races, fights 
with wild beasts, and gladiatorial games, for which troops 
of slaves are specially trained. If the lord goes to a prov- 
ince as governor or sojourns on one of his country estates, 
slave couriers and letter-carriers maintain daily communi- 
cation with the capital. And how shall we begin to tell of 
the slave retinue of the mistress, on which Bottiger has 
written a whole book (Sabina), and of the endlessly spe- 
cialized service for the care and education of the children! 
It was an incredible squandering of human energy that 
here took place. Lastly, by means of this many-armed 
organism of independent domestic economy, maintained 
as it was by a colossal system of breeding and training, the 
personal power of the slave-owner was increased a thou- 
sandfold, and this circumstance did much to render it pos- 
sible for a handful of aristocrats to gain control over half 
the world." 

The work of the State itself is not carried on otherwise. 
Both in Athens and Rome all subordinate officials and ser- 
vants are slaves. Slaves build the roads and aqueducts 
whose construction fell to the State, work in quarries and 
mines, and clean the sewers; slaves are the policemen, 
executioners and gaolers, the criers in public assemblies, 
the distributors of the public doles of corn, the attendants 



u 



Naturally this highly developed slave system is only to be found 
among the most wealthy class; but with similar conditions it recurs 
everywhere. Ellis, for example, says in his History of Madagascar, 
1, p. 194: "When a family has numerous slaves, some attend to cat- 
tle, others are employed in cultivating esculent roots, others collect 
fuel; and of the females, some are employed in spinning, weaving 
and making nets, washing and other domestic occupations." Even 
in the country of the Muata Yamwo, where, with the exception of 
smiths, there appear to have been no special craftsmen, the ruler had 
in his household his own musicians, fetich-doctors, smiths, hair- 
dressers, and female cooks. Pogge, Im Reiche d, Muata Jamwo, pp. 
231, 187. 



I02 THE RISE Of NATIONAL ECONOMY. 

of the colleges of priests in the temples and at sacrifices, 
the State treasurers, secretaries, the messengers of the 
magistrates; a retinue of public slaves accompanies every 
provincial officer or general to the scene of his duties. 
The means for their support came chiefly from the pubhc 
domains, the tributes of the provinces (in Athens, of the 
allies), of which Cicero says that they are qua^ pr(Bdia 
populi Romani; and finally, from contributions resembling 
fees. 

Similar fundamental features are presented by the eco- 
nomic life of the Latin and Germanic peoples in the early 
Middle Ages. Here, too, necessary economic progress 
leads to a further development of the autonomous house- 
hold economy, which found expression in those large hus- 
bandries worked with serfs and villeins upon the extensive 
landed possessions of the kings, the nobility, and the 
Church. In its details this manorial system has many points 
in common with the agricultural system of the later 
Roman Empire as developed by colonization. It has, also, 
considerable similarity with the centralized plantation sys- 
tem described above from the closing years of the Roman 
Republic. This rise of husbandry on a large scale with its 
subdivision of labour differs, however, in one important 
particular from the Roman. In Rome large estates engulf 
the small, and replace the arm of the peasant by that of 
the slave, who is later on transformed into the colonist. 
The economic advance involved in the extensive oikos 
husbandry had to be purchased by the proletarizing of the 
free peasant. In the manorial system of the Middle Ages 
the free owner of a small estate becomes, it is true, a vassal. 
But he is not ejected from possession; he preserves a cer- 
tain personal and economic independence, and, at the same 
time, shares in the fuller supply of goods which husbandry 



THE RISE OF NATIONAL ECONOMY. 103 

on a large scale provides under the system of independent 
domestic economy. 

How did this come about? 

In ancient Italy the small cultivator was ruined through 
his inability to support certain public burdens, especially 
military service, and because the pressure of war and 
famine drove him into the lamentable servitude of the 
debtor. In the Germanic and Latin countries of the Mid- 
dle Ages he placed his homestead for like reasons under 
the control of the large landed proprietor from whom he 
received protection and assistance in time of need. 

We can best understand the mediaeval manor by pic- 
turing to ourselves the economic life of a whole vil- 
lage as a unit with the manor-house its central point.^' 

''Though there were numerous villages whose inhabitants owed 
service to various proprietors, and many manors that included peas- 
ant holdings from various villages, yet the case here supposed must 
be regarded as the normal one. At the same time we must not for- 
get that most of the original evidence relating to these matters that 
we possess refers to the scattered possessions of the monasteries for 
which the manors formed the focal points, while for the estates of the 
great, and still more so for those of the smaller temporal proprietors 
in ancient times, we have scarcely any material at all. For these, 
however, our supposed case is to be regarded as normal in so far as 
the villages arose through a colony grouping itself about a single es- 
tate. For the purposes of our sketch we may also leave out of view 
the many distinctions in the legal position of those owing rent and 
service dues, especially the distinction between those belonging to the 
manor and those belonging to the mark. The latter, by virtue of 
the lord's supreme proprietorship over the common land, were also 
included in the economic system of the manor. Finally, I do not fail 
to appreciate the difference between the constitution of the villas of 
Charles the Great and the later administrative organization of the 
large Landowners, though I am of the opinion that the latter has only 
superficial points of contact with the economic life of the individual 
farm. For all further details I must refer the reader to Maurer, 
Gesch, d. Fronkofe; Inama-Stemegg, Die Ausbildung d, grossen Grund- 
herrschaften in Deutschland; and Lamprecht, Deutsches IVirtschafisUben 
im M. A.f especially I, pp. 719 ff. 



I04 THE RISE OF NATIONAL ECONOMY. 

Under this systenf the small landowner supervises in 
person, the large landowner through an overseer. The 
demesne land lying immediately about the manor-house 
is cultivated by serfs permanently attached to it, who there 
find food and lodging, and are employed in agricultural 
and industrial production, household duties, and the per- 
sonal service of the lord, under a many-sided division of 
labour. The demesne land is intermixed with the holdings 
of a larger or smaller number of unfree peasants, each of 
whom tills his hide of land independently, while all share 
with the lord the use of pasture, wood, and water. At 
the same time, however, every peasant-holding binds its 
occupant to perform certain services and to furnish cer- 
tain dues in natural products to the estate. These services 
consist of labour reckoned at first according to require- 
ment, later according to time, whether given in the fields 
at seed-time or harvest, on the pasture-land, in the vine- 
yard, garden or forest, or in the manorial workshops or 
the women's building where the daughters of the serfs are 
spinning, weaving, sewing, baking, brewing beer, etc. On 
the days devoted to manorial service the unfree labourers 
receive their meals at the manor-house just as do the 
manor-folk themselves. They are further bound to keep 
in repair the enclosures about the manor-house and its 
fields, to keep watch over the house, and to undertake the 
carrying of messages and the transport of goods. The 
dues in kind to be paid to the estate are partly agricultural 
products, such as grain of all kinds, wool, flax, honey, wax, 
wine, cattle, hogs, fowl, or eggs; j>artly wood cut in the 
forests of the mark and made ready for use, such as fire- 
wood, timber, vine-stakes, torch-wood, shingles, staves 
and hoops; partly the products of industry, such as wool- 
len and linen cloth, stockings, shoes, bread, beer, casks, 
plates, dishes, goblets, iron, pots and knives. This pre- 



THE RISE OF NATIONAL ECONOMY. 105 

supposes alike among the serfs and those bound by feudal 
service a certain specialization of industry, that would of 
necessity hereditarily attach to the hides of land in ques- 
tion and prove advantageous tiot merely to the lord's es- 
tate, but also to the occupants of the hides in supplying 
commodities. Intermediate between service and rent are 
duties of various kinds, such as hauling manure from the 
peasant's farm to the fields of the lord, keeping cattle over 
winter, providing entertainment for the guests of the 
manor. On the other hand the lord renders economic as- 
sistance to the peasant by keeping breeding-stock, by 
establishing ferries, mills, and ovens for general use, by 
securing protection from violence and injustice to all, and 
by giving succour from his stores, in accordance with his 
pledge, when crops failed or other need arose. 

We have here a small economic organism quite suffi- 
cient unto itself, which avoids the rigid concentration of 
the Roman slave estates and employs slaves only to the 
extent necessary for the private husbandry of the landlord 
conceived in its strictest sense.^* For this reason it is able 
to secure to the general body of manorial labourers sep- 
arate agricultural establishments for their own domestic 
needs, and therewith a certain personal independence. 
This is an instance of small partial private estates within 
the economy of the independent household similar to that 
which occurs, though of course on a much smaller scale, 
within the zadmga oi the South Slavs when conjugal 
couples establish separate households.^ ^ When the man- 

" According to Lamprecht, I, p. 782, the field labour-services of 
the serfs were applied to the cultivation of the individual stretches of 
manorial land (Beunden) or balks [unploughed strips] in the com- 
mon land, while the manorial serfs were employed only for the cul- 
tivation of the demesne. 

"Comp. Laveleye, as above, p. 468. 



io6 THE RISE OF NATIONj4L ECONOMY. 

orial group coincides in membership with the people of a 
mark, the members are in a certain sense, owing to the 
regulations forbidding the alienation of land or mark 
servitudes to non-residents, economically shut off from 
their neighbours. Internal unity is realized by means of 
separate weights and measures, which, however, serve not 
for safeguarding trade, but for measuring the dues in kind 
coming to the lord. 

For we must always bear in mind that the economic 
relation of the lord to those attached to his land, however 
much it may be regarded from the general point of view 
of mtttual service, is entirely removed from the class of 
economic relations that arise from a system of exchange. 
Here there are no prices, no wages for labour, no land or 
house rent, no profits on capital, and accordingly neither 
entrepreneurs nor wage-workers. We have in this case 
peculiar economic processes and phenomena to which his- 
torical political economy must not do violence, after such 
frequent complaints of harsh treatment in the past at the 
hands of jurisprudence. 

The surpluses of the manorial husbandry are the prop- 
erty of the lord. They consist entirely of goods for 
consumption which cannot be long stored up or turned 
into capital. On the estates of the king they are devoted 
as a rule to supplying the needs of the royal household, 
and the king, travelling with his retinue from castle to 
castle, claims them in person; while the large landed pro- 
prietors among the religious corporations and the higher 
nobility have them forwarded by a well-organized trans- 
port of their villeins to their chief seats, where as a rule 
they are likewise consumed. 

Thus in this economic system we have many of the phe- 
nomena of commerce, such as weights and measures, the 
carriage of persons, news, and goods, hostelries, and the 



THE RISE OF NATIONAL ECONOMY. 107 

transference of goods and services. In all, however, there 
is lacking the characteristic feature of economic exchange, 
namely, the direct connection of each single service with 
its reciprocal service, and the freedom of action on the 
part of the individual units carrying on trade with one 
another. 

But it matters not to what extent independent house- 
hold economy may be developed through the introduction 
of slave or villein labour, it will never succeed even in its 
highest development, to say nothing of its less perfect 
forms, in adapting itself sufficiently to the needs of human 
society for all time. Here we have continuously unfilled 
gaps in supply, there surpluses which are not consumed on 
the estates producing them, or fixed instruments of pro- 
duction and skilled labour which cannot be fully utilized. 

Out of this state of things arise fresh commercial phe- 
nomena of a particular kind. The landlord, whose harvest 
has failed, borrows corn and straw from his neighbour 
until the next harvest, when he returns an equal quantity. 
The man reduced to distress through fire or the loss of his 
cattle is assisted by the others on the tacit understanding 
that he will show the like favour in the like event. If any- 
one has a particularly expert slave, he lends him to a 
neighbour, just as he would a horse, a vessel, or a ladder; 
in this case the slave is fed by the neighbour. The owner 
of a wine-press, a malt-kiln, or an oven allows his poorer 
fellow villager the temporary use of it, in return for which 
the latter, on occasion, makes a rake, helps at sheep-shear- 
ing, or runs some errand. It is a mutual rendering of 
assistance; and no one will think of classifying such occur- 
rences under the head of exchange.^® 

"On the social duty of lending among primitive peoples, comp. 
Kubary, Ethnogr. Beitrdge 8, Kenntnis d. Karolincn-Archipels, p. 163. 



io8 THE RISE OF NATIONAL ECONOMY. 

Finally, however, real exchange does appear. The transi- 
tion-stage is formed of such processes as the following: 
* the owner of slaves lends his neighbour a slave weaver or 
carpenter, and receives in return a quantity of wine or 
wood of which his neighbour has a surplus. Or the slave 
shoemaker or tailor, whose labour cannot be fully turned 
to account, is settled upon a holding, on the condition 
that he work each year a certain number of days at the 
manor. At times when he has no labour dues to pay and 
little to do on his own land, he gives his fellow villeins in 
their peasant houses the benefit of his skill, receiving from 
them his keep, and in addition a quantity of bread or bacon 
for his family. Formerly he was merely the servant of the 
manor; now he is successively the servant of all, but of 
each only for a short time.^^ At an early stage barter in 
kind, aiming at a mutual levelling of wants and surpluses, 
is also met with, as corn for wine, a horse for grain, a piece 
of linen cloth for a quantity of salt. This trading process 
expands owing to the limited occurrence of many natural 
products and to the localization of the production of goods 
for which there is a large demand; and if the various 
household establishments are small, and the adjoining dis- 
tricts markedly dissimilar in natural endowments, it may 
attain quite a development.^® Certain articles of this trade 
become, as has often been described, general mediums of 
exchange, such as skins, woollen goods, mats, cattle, arti- 

" On the corresponding conditions in Greece and Rome, comp. my 
accounts in the Handwort. d. Staatswiss. (2d ed.), IV, pp. 369-71. 

**To this circumstance is to be ascribed the relatively highly de- 
veloped weekly market trade of ancient Greece and of the negro 
countries of to-day; in Oceania the small size of the islands and the 
unequal development among their inhabitants of both household work 
and agriculture even calls forth in places an active maritime trade. 
Similarly is the oft-cited maritime commerce of the ancient Greeks to' 
be regarded. 



THE RISE OF NATIONAL ECONOMY. iop 

cles of adornment, and finally the precious metals. Money 
comes into existence, markets and peddling trade arise; 
the beginnings of buying and selling on credit appear. 

But all this affects only the surface of the independent 
household economy; and, though the literature on the 
early history of trade and of markets has hitherto been 
far from familiarizing us with a proper estimate of these 
things, yet it cannot be too strongly emphasized that 
neither among the peoples of ancient times nor in the early 
Middle Ages were the articles of daily use the subject of 
regular exchange. Rare natural products, and locally manu- 
factured goods of a high specific value form the few articles 
of commerce. If these become objects of general demand, 
as amber, metal implements, ceramic products, spices and 
ointments in ancient times, or wine, salt, dried fish, and 
woollen wares in the Middle Ages, then undertakings must 
arise aiming at the production of a surplus of these articles. 
This means that the other husbandries will produce be- 
yond their own immediate requirements the trade equiva- 
lents of those articles as do the northern peoples their 
skins and vadhmdl, and the modern Africans their wares 
of bark and cotton, their kola nuts and their bars of salt. 
Where the population concentrates in towns there may 
even come into being an active market trade in the neces- 
saries of life, as is seen in classic antiquity, and in many 
negro countries of to-day. In fact even the carrying on 
of industry and trade as a vocation is to a certain extent 
possible. 

Still this does not affect the inner structure of economic 
life. The labour of each separate household continues to 
receive its impulse and direction from the wants of its own 
members; it must itself produce what it can for the satis- 
faction of these wants. Its only regulator is utility. " That 



no THE RISE OF NATIONAL ECONOMY, 

landlord is a worthless fellow," says the elder Pliny, " who 
buys what his own husbandry can furnish him " ; and this 
principle held good for many centuries after. 

One must not be led away from a proper conception of 
this economic stage by the apparently extensive use of 
money in early historic times. Money is not merely a me- 
Hjirrp nf eychat^ gre, jt is also a measure of value, a medium 
forjnaking.pa^ments andjor storing up wealth. Payments 
must also constantly be made apart from trade, such as 
fines, tribute-money, fees, taxes, indemnities, gifts of hon- 
our or hospitality; and these are originally paid in products 
of one's own estate, as grain, dried meat, cloth, salt, cattle, 
and slaves, which pass directly into the household of the 
recipient. Accordingly all earlier forms of money, and 
for a long time the precious metals themselves, circulate 
in a form in which they can be used by the particular 
household either for the immediate satisfaction of its wants 
or for the acquisition by trade of other articles of consump- 
tion. Those of special stability of value are pre-eminently 
serviceable in the formation of a treasure. This is es- 
pecially true of the precious metals, which in time of pros- 
perity assumed the form of rude articles of adornment, and 
as quickly lost it in time of adversity. Finally, it is man- 
ifest that the office of a measure of value can be performed 
by metal money even when sales are actually made in terms 
of other commodities, as is shown by the use in ancient 
Egypt of utetiy a piece of wound copper wire according to 
which prices were fixed, while payment was made in the 
greatest variety of needful articles.^® This is also shown 
by numerous mediaeval records in w^hich, far beyond the 
epoch here under review, prices are fixed partly in money 
and partly in horses, dogs, wine, grain, etc., or the pur- 

^ Erman, Aegypten u. dgypt, Leben im AUertum, pp. 179, 657. 



THE RISE OF NATIONAL ECONOMY. m 

chaser is left at liberty to make a money payment " in what 
he can " {in quo potxierif)}^ 

Lamprecht, discussing economic life in France in the 
eleventh century, affirms that purchases were made only x 
in cases of want;^^ the same holds in the main for sales 
as well. Exchange is an element foreign to independenTl 
household economy, and its entrance was resisted as long/ \ 
and as stubbornly as possible. Purchase always means 
purchase with immediate payment, and it is attended with 
solemn and cumbrous formalities. The earliest municipal 
law of Rome prescribes that the purchase must take 
place before five adult Roman citizens as witnesses. The 
rough copper that measures the price is weighed out 
to the seller by a trained weigh-master (libripens), while the 
purchaser makes a solemn declaration as he takes posses- 
sion of the purchased article. Contrasting with this the 
formal minuteness of early German trade laws, we are eas- 
ily convinced that in the economic period which witnessed 
the creation of this rigid legal formalism buying and sell- 
ing, and the renting of land or house, could not be every- 
day affairs. Exchange value accordingly exercised no| 
deep or decisive influence on the internal economy of the 
separate household. The latter knew only production for 
its own requirements; or, when such production fell short, 
the practice of making gifts with the expectation of re- 
ceiving others in return, of borrowing needful articles and 

" Under similar circumstances the same is true to-day. " Through- 
out West, Central, and East Africa quite definite and often quite com- 
plex standards for the exchange of goods have been formed, just as 
among ourselves, but with this difference, that coined money is gen- 
erally wanting. This, however, by no means prevents the existence 
of a system of intermediate values, though it be but as notions and 
names." — Buchner, Kamerun, p. 93. 

^Fransos, Wirtschaftskhen, p. 132. Comp. further his Deuisches 
WirischaftsUhen im M. A,, II, pp. 374 ff. 



112 THE RISE OF NATIONAL ECONOMY. 

implements, and, if need be, of plundering. The develop- 
ment of hospitality, the legitimizing of begging, the union 
of nomadic life and early sea-trade with robbery, the ex- 
traordinary prevalence of raids on field and cattle among 
primitive agricultural peoples, are accordingly the usual 
concomitants of the independent household economy. 

From what has been said it will be clear that under this 
^method of satisfying needs the fundamental economic phe- 
nomena must be dissimilar to those of modem national 
economy. Wants, labour, production, means of produc- 
tion, product, stores for use, value in use, consumption — 
these few notions exhaust the circle of economic phenom- 
ena in the regular course of things. As there is no social 
division of labour, there are consequently no professional 
classes, no industrial establishments, no capital in the sense 
of a store of goods devoted to acquisitive purposes. Our 
' classification of capital into business and trade capital, loan 
and consumption capital, is entirely excluded. If, conform- 
ably to widely accepted usage, the expression capital iz 
restricted to means of production, then it must in any case 
be limited to tools and implements, the so-called fixed 
capital. What modem theorists usually designate circu- 
lating capital is in the independent household economy 
merely a store of consumption goods in process of prep- 
aration, unfinished or half-finished products. In the regu- 
lar course of affairs, moreover, there are no sale-goods, no 
price, no circulation of commodities, no distribution of 
income, and, therefore, no labour wages, no earnings of 
management, and no interest as particular varieties of in- 
come.^^ Rent alone begins to differentiate itself from the 

" For most of the conceptions here mentioned there are no ex- 
pressions in Greek or Latin. They must be expressed by circumlo- 
cutions or by very general terms. This is true, in the first instance, 
of the conception income itself. The Latin reditus denotes the returns 
from the land. Tacitus makes use of a similar liberty when (Ann,, 



THE RISE OF N/ITIONAL ECONOMY. 113 

return from the soil, still appearing, however, only in com- 
bination with other elements of income^ aa-m«# fc|>^ ^'^ 

Perhaps, indeed, it is improper at this stage to speak of 
income at all What we call income is normally the fruit 
of commerce; in independent domestic economy it is the 
sum of the consumption goods produced, the gross return. 
This return, however, is all the more inseparable from 
general wealth the more the subjection of the husbandry 
to the hazard of the elements compels the accumulation of 
a store of goods. Income and wealth form indistinguish- 
able parts of a whole, one part of which is ever moving 
upward towards availability for use, another part down- 
ward to consumption, while a third is Rtnr<>f ^ up jn chest 
a nd box^ in cellar or storehouse, as a kind of assurance 
fund. 

To the last belongs mone y. In so far as it is used in 
tradeil irior the recipient as a rule not a provisional but a 
final equivalent It plays its chief part not as an intermedi- 
ary of exchange, but as a store of value and as a means of 
measuring and transferring values. Loans from one eco- 
nomic unit to another do indeed take place; but as a rule 
they bear no interest, and are made only for purposes of 
consumption. Productive credit is incompatible with this 
economic system. Where money-lending on interest in- 
trudes itself it appears unnatural, and, as we know from 
Greek and Roman history, is ultimately ruinous to the 
debtor. The canonical prohibition of usury thus had its 
origin not in moral or theological inclination, but in eco- 
nomic necessity. 

IV, 6, 3) he designates the revenues of the state as fructus publici. 
Compare with this the numerous and finely distinguished expressions 
for the conception wealth. Merces means not only wages, but also 
land-rent, house-rent, interest, price. So also the Greek twre^s. 
For the expressions vocation, occupation, undertaking, industry, 
neither of the classic languages has corresponding terms. 




J 



114 THE RISE OF NATIONAL ECONOMY. 

 

Where a direct state tax arose, it was regularly a tax on 
wealth, generally a species of land-tax. Such was the 
Athenian eiaipopa, the Roman tributum civium, and the 
scot or the bede of the Middle Ages. Along with these 
demand was made upon the wealth of the individual for 
direct services to the State or community, such as the 
furnishing of ships, the institution of festivals and enter- 
tainments (liturgies). The idea o f taxing income, how- 
ever natural and self -evident it may appear to usT^would 
h ave been simply imy »nreivah1e to our ancestors. 

rBy a process extending over centuries this independent 
household economy is transformed into the system of direct 
exchange; in the place of production solely for domestic 
use steps custom production. We have designated this 
stage town economy, because it reached its typical develop- 
ment in the towns of the Germanic and Latin countries 
during the Middle Ages. Still it must not be forgotten 
that even in ancient times beginnings of such a develop- 
ment are perceptible, and that at a later date they also 
appeared in the more advanced Slavic countries, albeit in 
considerably divergent form. 

The transition to this economic stage is seen at the 
' stage of domestic economy itself in the loss by the sep- 
arate household, founded upon the cultivation of the soil, 
of a part of its independence through inability longer to 
satisfy all its needs with its own labour, and through the 
necessity of permanent and regular reinforcement from 
the products of other estates. Yet there do not spring up 
at once establishments independent of the soil, whose 
members would derive their income entirely from the 
working up of industrial commodities for others, or the 
professional performance of services, or the conducting 
of exchange. On the contrary, each proprietor still seeks, 
as far as possible, to gain his livelihood from the land; if 



THE RISE OF N/ITION/IL ECONOMY. 115 

his wants go beyond this, he calls into requisition any 
special manual skill he may possess, or any particular pro- 
ductive advantage of his district, whether in field, forest, 
or water, in order to produce a surplus of some particular 
article. One will produce grain, another wine, a third salt, 
a fourth fish, a fifth linen or some other product of domes- 
tic industry. In this manner separate establishments come 
into existence specially developed in some one direction, 
and dependent upon a regular, reciprocal barter of their 
surplus products. This exchange does not at first demand 
an organized system of trade. But it does require more 
flexible commercial methods than were offered by the 
early laws. These are furnished by markets which still 
arise, in the main, under the household system. 

A market is the coming together of a large number of 
buyers and sellers in a definite place and at a definite time. 
Whether this occur in connection with religious feasts and 
other popular gatherings, or whether it owes its origin to 
the favourable commercial situation of a locality, it is al- 
ways an opportunity for producer and consumer to meet 
with their mutual trade requirements; and such in its gen- 
eral features it has remained down to the present day._ 
Markets and fixed trade are mutually exclusive. Where 
a merchant class exists, no markets are needed; where 
there are markets, merchants are superfluous. Only in 
cases where a country must import articles for which there 
is a demand and which it does not itself produce can there 
be developed at the early stage of household economy a 
distinct though not v^ry numerous class, uniting under 
their control the purchase, transport, and sale of these 
goods, and utilizing for this last purpose the trade oppor- 
tunities presented by the markets^ 

What changes, then, were wrought in this condition of 
things by the mediaeval town, and in what does the eco- 



Il6 THE RISE OF NyiTIONyiL ECONOMY. 

nomic system which we have designated as exclusive town 
economy consist? 

The mediaeval town is above all things a burg, that is, 
a place fortified with walls and moats which serves as a ref- 
uge and shelter for the inhabitants of the unprotected 
places round about. Every town thus presupposes the 
existence of a defensive union which forms the rural set- 
tlements lying within a greater or narrower radius into a 
sort of military community with definite rights and duties. 
It devolves upon all the places belonging to this commun- 
ity to cooperate in maintaining intact the town fortifica- 
tions by furnishing workmen and horses, and in time of 
war in defending them with their arms. In return they 
have the right, whenever occasi^ arises, to shelter them- 
selves, their wives and children, their*battle and movables, 
within its walls. This right is called the jright of burgess, 
and he who enjoys it is a burgher (burgensis). 

Originally the permanent inhabitants of the town differ 
in nowise, not even in their occupations, from those living 
in the rural hamlets. Like the latter they follow farming 
and cattle-raising; they use wood, water, and pasture in 
common; their dwellings, as may still be seen in the 
structural arrangement of many old cities, are farmhouses 
with bams and stables and large yards between. But their 
communal life is not exhausted in the regulation of com- 
mon pasturage and other agricultural interests. They are, 
so to speak, a permanent garrison stationed in the burg, 
and perform in rotation the daily watch-service on tower 
and at gate. Whoever wishes to settle permanently in the 
town must therefore not only be possessed of land, or a 
house at least ; he must also be provided with weapons and 
armour. 

The sentinel service and the extensive area of the 
town rendered necessary by the law of burgess demanded 



THE RISE OF NATIONAL ECONOMY. 117 

• 

a great number of men; and soon the town limits no 
longer sufficed for their maintenance. Then it was that 
the one-sided development of the household establish- 
ments, already described, lent its influence, and the town 
became the seat of the industries and of the markets as 
well. In the latter the country peasant continued to dis- 
pose of his surplus supplies, obtaining from the townsman 
that which he himself could no longer provide and which 
the latter now exclusively or almost exclusively produced, 
namely, industrial products. 

The burgess rights underwent a consequent extension. 
All who enjoyed them were exempt from market dues and 
town tolls. The right of free purchase and sale in the town 
market is thus in its origin an emanation from the rights 
of burgess. In this way the military defensive union be- ) 
came a territorial economic community based upon mutual I 
and direct exchange of agricultural and industrial products | 
by the respective producers and consumers. 

All market traders on their way to and from a market 
enjoyed — doubtless also in the period previous to the rise 
of towns — a particularly active royal protection, which was 
further extended to the market itself and to the whole 
market-town. The effects of this market-peaCe were to 
secure the market tradesmen during the time of their so- 
journ in the town against leg^l prosecution for debts previ- 
ously incurred, and to visit injuries inflicted upon their 
property or person with doubly severe punishment as be- 
ing extraordinary breaches of the peace. The market 
tradesmen are commonly known as Kaufleute, ntercatores, 
negotiatores, emptores}^ 

" Recent literature relating to the origin of the constitution of Ger- 
man towns has overlooked the very wide significance of the word 
Kaufmann and imagined that the innumerable towns existing within 
the German Empire towards the close of the Middle Ages, from 



Ii8 THE RISE OF NATIONAL ECONOMY. 

Inasmuch as the town inhabitants were themselves pecu- 
liarly dependent upon the market for their buying and 

Cologne and Augsburg down to Medebach and Radolfzell, were in- 
habited by merchants in the modern sense of the term, that is, by a 
specialized class of professional tradesmen, who are as a rule still 
represented as wholesale merchants. All economic history revolts 
against such a conception. What did these people deal in, and in 
what did they make payment for their wares? Besides, the very terms 
•used are opposed to it. The most prominent characteristic of the 
professional merchant in his relation to the public is not his custom 
of buying, but of sellingj Yet the chapman (Kaufmann) of the Mid- 
dle Ages is named from the word for buying — kaufen. In the State 
records of Otto III. for Dortmund from 990 to 1000 a.d. the emptores 
TrotmannuBj whose municipal laws, like those of Cologne and Mainz, 
are said to serve as' a model for other cities, are spoken of in the 
same connection as mercatores or negotiatores in other records. If the 
abbot of Reichenau in the year 1075 can with a stroke of the pen 
transform the peasants of Allensbach and their descendants into mer- 
chants («/ ipsi et eorum posteri sint mercatores) ^ no possible ingenuity 
of interpretation can explain this if we have in mind professional 
tradesmen. That in point of fact merchant meant any man who sold 
wares in the market, no matter whether he himself had produced them 
or bought the greater part of them, is evident, for example, from an 
unprinted declaration of the Council of Frankfurt in 1420 regarding 
the toll called Marktrecht (in Book No. 3 of the Municipal Archives, 
Fol. 80). There we find at the beginning that this toll is to be 
paid by " every merchant who stands on the street with his merchan- 
dise, whatsoever it be." Then follow, specified in detail, the in- 
dividual " merchants " or the " merchandise " affected by this toll. 
From the lengthy list the following instances may be given: dealers 
in old clothes, pastry-books, food-vendors, rope-makers, hazelnut- 
sellers, egg- and cheese-sellers with their carts, poultry-vendors who 
carry about their baskets on their backs, strangers having in their 
possession more than a malter of cheese, cobblers, money-changers, 
bakers who use the market-stalls, strangers with bread-carts, geese, 
wagons of vitch (fodder), straw, hay, cabbages, all vendors of linen, 
flax, hemp, yarn, who sell their wares upon the street. Here we have 
a confused medley of small tradesmen of the town, artisans and peas- 
ants. That buyers as well as sellers on the market were designated as 
KauHeute (merchants) is evident from numerous records; in fact, pas- 
sages might be cited in which, when the merchant is spoken of, it is 
the buyer that seems to be chiefly meant. 



THE RISE OF Nj4T!ONAL ECONOMY. 119 

selling, the specific name of market people or merchants 
was more and more applied to them as the importance of 
the market as their source of supply increased. Propor- 
tionately with this change, however, the region from which 
this market drew its supplies and to which it sold extended 
farther into the country. No longer did it coincide with 
the domain of burgess rights, whose importance for the 
rural population must of itself have diminished with the 
increasing security of the whole country against external 
attack. On the other hand, with the growth of the indus- 
tries the whole town, and not merely the space originally 
set apart for the exclusive purpose, became the market; 
market-peace became town-peace, and for the maintenance 
of the latter the town was separated from the general state 
administration as a special judicial district. " City air 
makes free " became a principle. Thus arose a social 
and legal gulf between burgher and peasant which the 
thirteenth and fourteenth centuries vainly sought to bridge 
over by an extramural and intramural citizenship. The 
name burgher was finally restricted to the members of the 
community settled within the town limits; and the times 
lent to this title a legal and moral significance in which the 
state idea of the ancient Greeks appeared to have returned 
to life. 

We cannot here occupy ourselves further either with the 
development of the municipal constitution and its self- 
administration based upon corporative gradations, or with 
the political power which the towns of Germany, France, 
and Italy obtained in the later Middle Ages. We have to 
do only with the matured economic organization of which 
these towns formed the central points. 

If we take a map of the old German Empire and mark 
upon it the places that, up to the close of the Middle Ages, 
had received grants of municipal rights — ^there were prob- 



I20 THE RISE OF NATIONAL ECONOMY. 

ably some three thousand of them — ^we see the country 
dotted with towns at an average distance of four to five 
hours' journey in the south and west, and in the north and 
east of seven to eight. All were not of equal importance; 
but the majority of them in their time were, or at least en- 
deavoured to be, the economic centres for their territory, 
leading just as independent an existence as the manor be- 
fore them. In order to form a conception of the size of 
these districts, let us imagine the whole country evenly 
divided among the existing municipalities. In this way 
each town in southwestern Germany has on the average 
forty to somewhat over fifty square miles, in the central 
and northeastern parts between sixty and eighty-five, and 
in the eastern from somewhat over one hundred to one 
hundred and seventy. Let us imagine the town as always 
situated in the centre of such a section of country, and it 
becomes plain that in almost every part of Germany the 
peasant from the most distant rural settlement was able 
to reach the town market in one day, and be home again 
by nightfall.^^ 

The whole body of municipal market law, as formulated 
in early times by the lords of the town and later by the 

** Although since the Middle Ages many places have lost their town 
franchises, while others have gained them for the first time, yet the 
number of places that to-day bear the name of town (Stadt) furnishes 
a pretty correct idea of what it then was. There is in Baden at pres- 
ent one city to every 132 square kilometres of territory [i sq. km. = 
about I sq. mile], in Wtirtemberg to 134, in Alsace-Lorraine to 137, 
in Hesse to 118, in the kingdom of Saxony to 105, in Hesse-Nassau 
to 145, in the Rhine Province to 193, in Westphalia to 196, in the 
province of Saxony to 175, in Brandenburg to 291, in the kingdom of 
Bavaria to 3i8, in Hanover to 341, in Schleswig-Holstein to 350, in 
Pomerania to 412, in West Prussia to 473, and in East Prussia to 552. 
The fever for founding municipalities, which racked many mediaeval 
rulers, called into existence a multitude of towns that lacked vitality. 
Well known is the prohibition in the Sachsenspiegel that " No market 
shall be founded within a mile of another." Weiske, III, 66, 8 i. 



THE RISE OF NATIONAL ECONOMY. 12 1 

town councillors, is summed up in the two principles, that, 
as far as at all possible, sales must be public and at first hand, 
and that everything which can be produced within tlie town/ 
itself shall be produced there. For products of local manu- 
facture intermediary trade was forbidden to everyone, even 
to the artisans; it was permitted with imported goods only 
when they had already been vainly offered on the market. 
The constant aim was to meet amply and at a just price 
the wants of the home consumers, and to give full satis- 
faction to the foreign customers of local industry. 

The territory from which supplies were drawn for the 
town market, and that to which it furnished commodities, 
was identical. The inhabitants of the country brought in 
victuals and raw materials, and with what they realized 
paid for the labour of the town craftsmen, either in the 
direct form of wage-work or in the indirect form of finished 
products, which had been previously ordered or were se- 
lected in the open market from the artisan's stand. 
Burgher and peasant thus stood in the relationship of mu- 
tual customers: what the one produced the other always 
needed; and a large i>art of this exchange trade was per- 
formed without the mediation of money, or in such a way 
that money was introduced only to adjust differences in 
value. 

Town handicraft had an exclusive right of sale on the I 
market. The productions of other places were admitted I 
only when the industry in question had no representatives 
within the town. They were usually offered for sale by the 
foreign producers at the annual fairs; at this one point 
the spheres of the various town markets overlap. But 
even here the most essential feature, the direct sale by 
producer to consumer, is also observed, though only in 
exceptional instances. If a trade capable of supporting a 
craftsman was not represented in the town, the council 



\ • 



122 THE RISE OF NMTIONML ECONOMY. 

called in a skilled master workman from outside and in- 
duced him to settle by exemption from taxation and other 
privileges. If he required considerable initial capital, the 
town itself came to his aid, and at its own expense built 
work- and sale-shops and established mills, g^inding- 
works, cloth-frames, bleaching-places, dye-houses, fulling- 
mills, etc., — ^all with a view to satisfying the greatest pos- 
sible variety of wants by home production^ 

Although direct dealing with the consumer of his 
wares ^' tended necessarily to keep alive in the artisan a 
sense of personal responsibility, an effort was made to 
brace this moral relationship by special ordinances. Hand- 
work is an ofHce that must be administered for the general 
welfare. The master shall furnish " honest " work. So far 
as the personal services of the craftsman remained avail- 
able to his customers, a regular rate was fixed governing 
the amount he could claim in wages and board while on his 
itinerancy. In cases where the customer furnished him 
with the raw material in his own home, as, for instance, tin 
to the pewterer, silver and gold to the goldsmith, or yarn 
to the weaver, provision was made that it should not be 
adulterated. Where, on the contrary, the artisan supplied 
the material there were erected in the market, about the 
churches, at the town gates, or in particular streets, public 
sale-booths which often served also as work-shops (bread- 
stands, meat-stalls, drapers' and cloth shops, furriers' 
booths, shoemakers' benches, etc.). It was a market rule 
that those vending the same wares should do their selling 
alongside one another in open and mutual competition 
and under the supervision of the market wardens and over- 

" Here and there this was further secured by the regulation that 
not even the wife of the craftsman might represent him in selling. 
Comp. Gramich, Verf. u. Verw. d. St Wursburg vom, XIII. bis XV. 
Jhdt., pp. 38 f. 



THE RISE OF NATIONAL ECONOMY. 123 

seers, and this rule was extended to craftsmen who merely 
worked at home on orders, in that for the most part they 
lived side by side on the same street. Many cities have 
preserved to the present day the remembrance of this con- 
dition of things in the names of their streets (such as Shoe- 
maker, Turner, Weaver, Cooper, Butcher, Fisher Streets), 
many of which led directly into the old market square. In 
this way the greatest part of the town, or even the whole of 
it, bore the outward aspect of one large market. It 
is well known that the many prescriptions regarding the 
raw material to be used, the method of doing work, the 
length and breadth of cloths, and the direct regulation of 
prices must have served for the protection of the con- 
sumer.^* 

Just as the urban craftsman enjoyed within the town 
and the extramural judicial district (Bannmeile) the ex- 
clusive right of selling the products of his handicraft, so 
the urban consumer possessed for the same area the ex- 
clusive right to purchase imported commodities. This 
right can be exercised, to be sure, only when the imported 
goods actually come to market and stand on sale for the 
proper length of time. To effect this a law of staple is 
introduced; foreselling in the country places or before tKe\ 
town gates is forbidden; selling to middlemen, artisans, i 
and strangers is permitted only after the consumers are \ 
supplied, and then usually with the limitation that the lat- 
ter, if they so wish, may have a share; and lastly, the 
withdrawing of goods once brought to market was for- 
bidden, or permitted only after they had remained three 
days unsold.*^ 

*• For the sake of brevity we refer for all details in this connection 
to Stieda in the Jhb. f. N.-6k. u. Statistik, XXVII, pp. 91 flF. 

"These ordinances were most carefully wrought out for the com 
trade. See Schmoller, Jhb. f. Gcsetzg. Verw. u. Volksw., XX, pp. 
708 flf. 



124 THE RISE OF NATIONAL ECONOMY. 

\ But against the foreign seller there always prevails a 
deep-rooted mistrust. To this is due the existence of that 
peculiar system of exchange through official interme- 
diaries, measurers, and weighers. To-day the State con- 
trols weights and measures by official standards and public 
inspections, and leaves the ttTi^ toltt^ 'buyers aCnd seHers 
themselves. In the Middle Ages the technical means for 
constructing exact measures and ensiMng*.their:accuracy 
were wanting. Common field-stones — ^and at the Frank- 
furt fairs as late as the fifteenth century even wooden 
blocks — ^were used as weights. In order, however, to de- 
termine accurately the amount of goods exchanged, the 
handling of the measures was withdrawn from the parties 
themselves and entrusted to special officers, whose pres- 
ence was made obligatory at every sale made by an out- 
sider. It was the duty of these intermediaries to bring 
buyer and seller together, to assist in fixing the price, to 
test the goods for possible defects, to select for the pur- 
chaser the quantity he had bought, and to see to its proper 
delivery. The intermediary was forbidden to trade for 
himself; he was not even allowed at the departure of the 
foreign tradesman, whom he generally lodged, to purchase 
remnants of goods remaining unsold. 

This system of direct exchange is found, though with 
many local peculiarities, carried out to the most min- 
ute details in all mediaeval towns. This means that the 
actual circumstances in which its principles were devel- 
oped render it inevitable. How far it was really prac- 
ticable can only be decided when we are able to determine 
what proportions trade assumed under it. 

It is beyond question that a retail trade had taken root 
in the towns. To it belonged all who " sell pennyworths 
for the poor man." To understand this, we must bear in 
mind that all well-to-do townspeople were accustomed 



THE RISE OF NATIONAL ECONOMY. "S 

to purchase their supplies directly from foreign merchants 
at the weekly and yearly markets. The poor man was 
unable to make provision for any length of time; he lived^ 
as he does to-day, " from hand to mouth." For him the 
retail tradesman, accordingly, undertook the keeping of 
stores for daily sale. 

We can distinguish three groups of such small trades- 
men, namely, grocers, peddlers, and cloth-dealers. In the 
earlier half of the period of town economy the last were 
the most important, as in many towns there was -no local 
wool-weaving done. With its development their activity 
was limited to the handling of the finer kinds of Dutch 
cloths, silks, and cottons, or else they made room for the 
weavers in their shops. 

The wholesale trade was exclusively itinerant and market 
or fair trade; and down to the close of the Middle Ages 
the majority of the towns probably saw no merchants 
settled within their walls who carried on wholesale trade 
from permanent headquarters. Only commodities not pro- 
duced within the more or less extensive district from 
which a town drew its supplies were the subject of whole- 
sale trade. We know of but five kinds: (i) spices and 
southern fruits, (2) dried and salted fish, which were then 
a staple food of the people, (3) furs, (4) fine cloths, (5) for 
the North German towns, wine. In certain parts of Ger- 
many salt would also have to be included. In most cases, 
however, the town council ordered it in large quantities 
directly from the places of production, stored it in the 
municipal salt warehouses, and after a monopolistic ad- 
vance of its price gave it out to be disposed of by pe ddler 
and saltman, who paid a fee for the privilege. I Usually the 
foreign wholesale dealers^'' were permitted to sell their 

"In Jhb. f. Nat-Ok. u. Stat, 3 F., XX (1900), pp. i ff., G. v. Below 
attempts to prove that in the Middle Ages there existed no class of 



126 THE RISE OF NATIONAL ECONOMY. 

wares only in large lots or in minimum quantities — ^in the 
case of spices, for example, not under 12^ pounds. The local 
retailers and peddlers then carried on the sale in detail. 
This also holds good for many large producers, as, for 
instance, hammersmiths, who might sell to founders the 
iron they had failed to persuade smiths and private indi- 
viduals to buy. 

Though the limits of the territory from which the mar- 
ket of a mediaeval town drew its supplies and to which its 
sales were made cannot be determined with precision, see- 
ing that they varied for different wares, yet from the eco- 
nomic point of view they formed none the less an inde- 
pendent region. Each town with its surrounding country 
constituted an autonomous economic unit within which its 
whole course of economic life was on an independent foot- 
ing. This independence is based upon special currency, 
and special weights and measures for each locality. FTfie 
relation between town and country is as a matter of fact 
a compulsory relation such as that between the head and 
the limbs of the body, and it displays strong tendencies to 
assume the forms of legalized compulsion. The extra- 
mural jurisdiction of the town, the prohibitions of export 
and import already met with, the differential tolls, and the 
direct acquisition of territory on the part of the large towns 
plainly point in that directiony 

Many as are the objections which may be urged against 
deducing the constitution of the town from that of the 
manor, the economic system of the town can be properly 

wholesale merchants, and that the characteristic feature of those times 
was for wholesale and retail trade to be carried on by the same person- 
He therefore takes exception to our use of the term wholesale mer- 
chant. I cannot, however, see why when these persons appeared as 
wholesalers they should not receive the name, particularly as there is 
no evidence to show that wholesale merchants were ever met with who 
were not at the same time retail traders. 



THE RISE OF NATIONAL ECONOMY. 127 

understood and explained only as a continuation of the 
manorial systen^ What existed in the latter in mere 
germs and beginnings grew into finished organisms and 
systems of organisms; factors that under the independent 
household economy were found grouped in primitive 
shapelessness, have now been differentiated by subdivision 
and made independent. The forced division of labour of 
the manor has broadened into a free division of production 
between peasants and burghers, displaying morever 
among the latter a varied multiplicity of separate trades. 
The manorial workman carrying on domestic labour has 
become the wage-earning craftsman, who in time comes 
to possess his own business capital in addition to his tools. 
The vital thread connecting manorial and cottier economy 
has at length been severed; the separate household estab- 
lishments have gained an independent existence; trade 
between them is no longer conducted on the basis of a gen- 
eral return, but on the basis of a specific payment for ser- 
vices given and received. To be sure they have not yet 
fully emancipated themselves from the soil, even in the 
town; production is still dependent in large measure upon 
the domestic husbandry; but there have been formed the 
distinct vocations of agriculturalist, artisan, and trades- 
man which have g^ven a specific direction to the activities 
and lives of those following them. Society has become 
differentiated; c/ayj^j now exist; and these were unknown 
before. 

The whole circle of economic life has gained in fulness 
and variety in comparison with the independent domestic 
economy; the membership of the separate household es- 
tablishments has become smaller; the individuals are inter- 
dependent; they undertake certain functions for each 
other; exchange value is already forcing its way as a deter- 
mining factor into their inner life. But the producing com- 



128 THE RISE OF NATIONAL ECONOMY. 

munity still coincides with the consuming community; the 
handicraftsman's assistants drawn from outside, and even 
the tradesmen, are members of their employer's house- 
hold, subject to his disciplinary control. He is their " mas- 
ter/' they are his " servants." 

And still, as ever, by far the greater proportion of com- 
modities does not pass beyond the lintels of the place of 
production. A small part finds its way by the process of 
exchange into other establishments, but the way travelled 
is a very short one, namely, from producer to consumer. 
There is no circulation of goods. The sole exceptions are 
the articles of foreign trade few in number, and of the 
petty retail trade. They alone became wares. They alone 
must frequently take on the form of money before reach- 
ing their domestic destination. But here we deal with an 
exception to the system of direct exchange, not with a con- 
stituent element of the whole economic order. 

Though at this early point there is a social division 
of labour, and also an interrelationship of trades, there are 
as yet neither fixed industrial undertakings nor the neces- 
sary industrial capital. At most we are justified in speak- 
ing of trade capital. Handicraft undertakes work, but it 
is no business undertaking. In the forms of itinerant 
handicraft and home work it almost totally lacks capital. 
It means wage-labour embodied in the material of another. 
Even where the craftsman works with his own tools the 
product does not increase in value from the constant in- 
corporation of fresh increments of capital, but from the fact 
that labour is being invested in it. 

The amount of loan and consumption capital is also ex- 
ceedingly small. It may even be doubted whether in medi- 
aeval trade credit operations can be spoken of at all. 
Early exchange is based upon ready payment; nothing is 
given except where a tendered equivalent can be directly 



THE RISE OF NATION j4L ECONOMY. 129 

received. Almost the entire credit system is clothed in 
the forms of purchase. This was the case with the heredi- 
tary peasant holding and with the giving of town building- 
sites in return for a ground-rent, in which instances the 
land was looked upon as the purchase-price for the right 
to levy rent.^® So it was also under the earlier law where 
the land placed at the disposal of the money-lender passes 
as a temporary equivalent into the keeping of the " cred- 
itor " and becomes his property if the debtor fails to repay 
the loan. Economically considered, this commercial act 
differs in no way from selling in order to buy again; and 
it is admitted that it is now scarcely possible to discover a 
legal distinction between the two. A similar character is 
borne by the most common of urban credit transactions, 
the purchase of rents or stocks,^® as the name itself indi- 
cates. The price is the capital loaned; the commodity 
exchanged is the right to draw a yearly rent, which the 
borrower on the security of a house transfers in such a 
manner that the owner for the time being receives the rent. 
The rent is of the nature of a charge upon the soil, and is 
for a long time incapable of release; the party responsi- 
ble answers for it with the house or land upon which it is 
fixed, but not with his other assets. It is thus a charge 
anly upon the real property that carries it, whose rentabil- 
ity it proportionately diminishes. The person entitled to 
the rent has absolutely resigned the purchase-price paid; 
the document conferring the right to draw the rent can 
be transferred without formality, just as a bill payable to 
bearer. Every personal relationship is thus eliminated 
from the whole transaction, which lacks that element of 

"Compare in connection with this whole section the luminous ex- 
planation by A. Hcusler, Institutionen d, deutsch. Privatrechts, II, pp. 
128 ff. 

^Renienkauf and GiiUkauf, 



I30 THE RISE OF NATIONAL ECONOMY. 

trust peculiar to credit. The right of redemption bears 
the same character; it is the sale of the rent under reserva- 
tion of the privilege of re-purchase. 

As in dealings in real property, so also with mov- 
ables, the credit transaction is but a variation of ready pay- 
ment. The security, as Heusler says, is a provisional 
transfer on the part of the debtor of an equivalent that is 
still redeemable (forfeitable security), not a covering of the 
debt, which may eventually be claimed by the creditor and 
realized upon by being converted into money (saleable 
security). The pawnbroking business of the Jews ®® is in 
fact similar to our modern sale with right of redemption, 
and the " goods credit " extended to-day by craftsmen 
and shopkeepers, takes in the Middle Ages the form of 
purchase on security.^^ If at the same time we con- 
sider that when personal credit was given in those times 
the debtor almost always had to agree to submit to the 
creditor's right of security; that in most instances he 
could get money only by furnishing the best security un- 
der pledges to the lender and similar burdensome condi- 
tions; that the creditor in addition reserved the right, in 
case of default, to obtain the money from Jews at the debt- 
or's expense; and that the fellow citizens or heirs of the 
foreign debtor could be distrained upon for the amount of 
his debt, — we see plainly that in the town economy of the 
Middle Ages a credit system in the modem sense cannot 
be spoken of.^^ 

•• Comp. my Bevolkerung von Frankfurt, I, pp. 573 flF. 

" Comp. the interesting examples in Stieda, as cited above, p. 104. 

"A striking resemblance to the mediaeval credit system is offered 
by the Greek system and its legal forms. Here likewise purchase and 
loan are largely synonymous terms, and the language has not arrived 
at the stage of distinguishing sharply the notions of buying, pledging, 
renting, and subjecting to conditions. The Greek mortgage laws 
coincide in all important points with the early German. Comp. K. F. 



THE RISE OF NATIONAL ECONOMY, 131 

There are two things in connection with this that must 
appear especially strange to a student of modem political 
economy, namely, the frequency with which immaterial 
things (relationships) become economic commodities and 
subjects of exchange, and their treatment under commer- 
cial law as real property. These show clearly how primi- 
tive exchange sought to enlarge the sphere denied it un- 
der the existing conditions of production by awkwardly 
transforming, into negotiable property, almost everything 
it could lay hold upon, and thus extending infinitely the 
domain of private law. What an endless variety of things 
in mediaeval times were lent, bestowed, sold, and pawned! 
— the sovereign power over territories and towns; county 
and bailiff's rights; jurisdiction over hundreds and can- 
tons; church dignities and patronages; suburban monop- 
oly rights; ferry and road privileges; prerogatives of mint- 
age and toll, of hunting and fishing; wood-cutting rights, 
tithes, statute labour, ground-rents, and revenues; in fact 
charges of every kind falling upon the land. Economi- 
cally considered, all these rights and "relationships" share 
with land the peculiarity that they cannot be removed 
from the place where they are enjoyed, and that they can- 
not be multiplied at will. 

Income and wealth are at this stage not yet clearly distin- 
guished from each other. When in Basel in the year 145 1 
the " new pound-toll " was introduced it was prescribed 
that it should be paid: (i) from the selling-price of wares, 

(2) from the capital invested in the purchase of rents, and 

(3) from the amount of rents received.^^ On every pound 

Hermann, Lehrbuch der griech. PrivataUertumer mit Einschluss der Rechts- 
altertumer, §§ 67 and 68. The old Roman Mucia and its later form as 
pignus may be mentioned for purpose of comparison. 

 Comp. Schonberg, Finansuerhaltnisse d, Stadt Basel im XIV. und 
XV. Jhdt., p. 267. 



132 THE RISE OF NATIONAL ECONOMY. 

four pfennigs were to be paid, no matter whether the com- 
modity had changed hands as purchase-money, as capital, 
or as interest. In the first instance we have to do, accord- 
ing to our terminology, with gross revenue, in the second 
with property, in the third with net income; and yet all 
three cases are treated alike. Similar examples might be 
adduced from the tax-regulations of other cities.'* 

Two of our modern classes of income, however, now 
come more clearly into view, namely, ground-rent and 
wages. The latter bears, to be sure, a peculiar character; 
it is handicraft-wage, compensation for the use of the 
craftsman's labour on behalf of the consumer, and not, as 
to-day, the price paid to the wage-worker by the en- 
trepreneur. Still this price already exists in germ in the 
slight money-wage that the artisan gives to his journey- 
man in addition to free maintenance, thus enabling the 
latter to supply independently a limited proportion of his 
wants. Earnings of management appear only in the 
sphere of trade, and thus, like it, are the exception; more- 
over through connection with transportation they are 
more coloured by elements of labour-wage than are the 
earnings from trade to-day. As a rule interest takes 
on the form of ground-rent; and the same is true of the 
many kinds of revenues arising from the juridical rela- 
tionships that enter into trade. As credit operations 
usually take the form of purchase, they almost always 
mean for the creditor the actual transfer of a portion of 
his property, in order that he may receive a yearly income 
or a continuous usufruct. This is a rule, for example, of 
enfeoffment; with mortgaged property, according to the 
early law, it involved the transfer of the natural yield of 

■* For details sec my paper on two mediaeval tax-regulations, KUinere 
Beitrdge s. Geschich, von DozetUen d. Leipsiger HochschuU. Festschrift s, 
dritt. Historikertage (Leipzig, 1894), pp. 123 ff. 



THE RISE OF NATIONAL ECONOMY, I33 

the land, and with rent-purchases, of the ground-rent or 
rent. On this basis arose the earliest type of personal in- 
surance and at the same time the chief form of public 
credit: the negotiating of annuities. 

Public economy is still mainly of a private character: 
revenues from domains, sovereign prerogatives, tithes, 
statute labour, services, ground-rents, and fees preponder- 
ate in the State, market revenues and imports on con- 
sumption in the towns. The general property-tax con- 
tinues to be the only direct tax, and mingled with it here 
and there are elements of an income-tax. It is indeed 
levied more frequently than in the preceding period, but 
still it is not regular. 

In Germany the economic supremacy of the towns over 
the surrounding country blossomed only in a few places 
into political sovereignty. In Italy a parallel development 
led to the formation of a tyranny of the cities; in France 
the beginnings of autonomy on the part of free municipal 
communities were early suppressed by the kings with the 
aid of the feudal nobility. The reason is that in Germany, 
as in France, everything that lay without the town wall 
was overlaid with a mass of feudal institutions. True the 
great landed proprietors had long since given up the per- 
sonal management of their manorial estates, — which be- 
came for the owner, just as the town land and house prop- 
erty for the patrician families, nothing more than a source 
of income. But their original economic power had now be- 
come political, the landed proprietors were now territorial 
princes, and in the course of this transformation there had 
arisen a new and widely ramified class of small titled pro- 
prietors whose interests, purely agrarian in character, were 
closely linked to those of the princes. Hence that keen 
struggle in Germany between burgher and noble which 



134 THE RISE OF NATIONAL ECONOMY. 

fills the closing centuries of the Middle Ages, and in which 
the towns maintain the political autonomy they had for 
the most part acquired from their lords by purchase and 
unredeemed pledges, though they fail to wrest the peasant 
class from the feudal powers. 

It can thus be said that the economic development of the 
towns in Germany and France remained incomplete; and 
that they did not accomplish what the most vigorous 
types of the period of autonomous household economy had 
actually achieved, namely, the transmuting of th^ir eco- 
nomic power into politi cal ^ independence. This was 
perhaps fortunate for us. \ In Italy the wealth of the 
cities expropriated in all directions the possessions of the 
peasant, and down to the present has continued to exploit 
him as a wretched metayer. In Germany the nobility were, 
indeed, able to make of him a feudatory; but the con- 
ception of nationality, which first came to life in the terri- 
torial sovereignties, served to prevent his proletarization. 

The final development of national economy is in its es- 
sence a fruit of the political centralization that begins at 
the close of the Middle Ages with the rise of territorial 
state organizations, and now finds its completion in the 
creation of the unified national State. Economic unifica- 
tion of forces goes hand in hand with the bowing of private 
political interests to the higher aims of the nation as a 
whole. 

In Germany it is the more powerful territorial princes, 
as opposed to the rural nobles and the towns, who seek 
to realize the modern national idea, often certainly under 
great difficulties, especially when their territories were 
widely scattered. From the second half of the fifteenth 
century we have many indications of a closer economic 
union, such as the creation of a territorial currency in 
place of the numerous town currencies, the issue of terri- 



THE RISE OF NATIONAL ECONOMY, 135 

torial regulations regarding trade, markets, industry, for- 
estry, mining, hunting, and fishing, the gradual formation 
of a system of sovereign prerogatives and concessions, the 
promulgation of territorial laws, conducive to greater legal 
unity, and the emergence of an ordered public economy. 

But for centuries longer agricultural interests predom- 
inate in Germany, and as against them the exertions of the 
imperial power in the direction of a national economic 
policy lamentably failed. On the other hand the Western 
European states, — Spain, Portugal, England, France, and 
the Netherlands, — from the sixteenth century on appear 
externally as economic units, developing a vigorous colo- 
nial policy in order to turn to account the rich resources of 
their newly acquired possessions over sea. 

In all these lands, though with varying degrees of se- 
verity, appears the struggle with the independent powers 
of the Middle Ages, — the greater nobility, the towns, the 
provinces, the religious and secular corporations. The 
immediate question, to be sure, was the annihilation of the 
independent territorial circles which blocked the way to 
political unification. But deep down beneath the move- 
ment leading to the development of princely absolutism, 
slumbers the universal idea that the greater tasks con- 
fronting modem civilization demanded an organized union 
of whole peoples, a grand living community of interests; 
and this could arise only upon the basis of common eco- 
nomic action. Each portion of the country, each section 
of the population, must in the service of the whole take 
over those duties that its natural endowments best fitted 
it to perform. A comprehensive partitioning of functions 
was necessary, a division into callings embracing the whole 
population; and this division itself presupposed a highly 
developed commerce and an active interchange of goods 
amongst the population. If the sole aim of all economic 



13^ THE RISE OF NATIONAL ECONOMY. 

effort was in ancient times to make the house autonomous 
in the satisfaction of its wants, and in later mediaeval times 
,to supply the needs of the town, there now comes into 
being an exceedingly complex and ingenious system for 
meeting the wants of the entire nation. 

The carrying out of this system is from the sixteenth to 
the eighteenth century the economic effort of all the ad- 
vanced European states. The measures employed to at- 
tain this object are modelled in almost every detail upon 
the economic policy of the mediaeval towns.^*^ They are 
generally summed up under the name of the mercantile 
system. This latter has long been regarded as a theoretical 
edifice culminating in the principle that the wealth of a 
country consists in the amount of coin within its borders. 
To-day, in all probability, this conception is universally 
abandoned. Mercantilism is no dead dogma, but the 
active practice of all leading statesmen from Charles V. 
to Frederick the Great. It found its typical development 
in the economic policy of Colbert. He sought the removal 
or reduction of the internal customs and tolls, the intro- 
duction of a unified customs system on the national bor- 
ders; the assuring to the country of a supply of the neces- 
sary raw materials and means of sustenance by hindrances 
to export, and the institution of the forest regalia. He 
fostered industry on a large scale by the establishment of 
new industrial branches with state support and technical 
supervision, by the exclusion of foreign competition 
through prohibitive tariffs, and by the building of roads, 
canals, and harbours. With the same end in view he strove 
to unify the system of weights and measures, and to regu- 
late commercial law and the commercial news-service. 

"For the German states this development is excellently portrayed 
by Schmoller in the Jhb. f. Gesetzgeb. Verw. u. Volksw. VIII (1884), 

pp. 22flF. 



THE RISE OF NATIONAL ECONOMY. 137 

The cultivation of the technical arts, fine arts, and science 
in special state institutions ; the systematizing of state and 
communal expenditure, the removal of inequalities in taxa- 
tion, also served his one purpose — to create an independent 
national economy vfhich should satisfy all the needs of the citi- 
zens of France by national labour, and by an active internal 
trade bring all the natural resources of the country and all 
the separate forces of the people into the service of the 
whole. In considering the special encouragement given 
by " Colbertism " to foreign commerce, the marine, and 
colonial trade, it has all too often been overlooked that 
these measures also strengthened the inner resources of 
the country, and that the theory of the balance of tradq 
became a necessity at a time when the transition from 
the still predominant household production to the system 
of universal exchange -indispensably postulated the in- 
crease of the monetary medium of circulation. 

Along with the state measures we must not fail to take 
account gf the social forces working in the same direction. 
These naturally had their starting-point in the towns. 
Here, by a gradual process of transformation, loaning at ' 
interest had been evolved from the purchase of rents; and 
thus in the course of the sixteenth century a true credit 
system arose. In this we may see the influence of whole- 
sale trade, which first discovered the secret of making 
money with money. Through the liberation of capi- ' 
tal invested in rents the wealth of the rich townsmen 
acquired a greatly increased mobility and accumulative 
power. Loan capital now took its place at the side of 
trade capital, hitherto the only kind of capital; and the 
two supplemented and supported each other in their fur- 
ther development. 

The immediate result was a notable expansion of trade. 
Certain towns began to rear their heads above the uniform 






138 THE RISE OF NATIONAL ECONOMY, 

mass of mediaeval market and handicraft towns as centres 
of state administration or as emporiums of trade. In Ger- 
many, which through the decline of the Hansa and the 
change in the highways of the world's trade, had lost most 
of its importance as an intermediary of trade with the 
North, the change manifests itself to some extent in the 
growing importance of the great fairs, and in the decad- 
ence of the local markets. The Frankfurt fair reached its 
zenith in the sixteenth century, that of Leipzig consider- 
ably later. But soon trade capital is no longer content 
with the importation and handling of foreign products; 
it becomes, for native industry and the surplus products 
of the peasant's domestic labour, commission capital. 
Wholesale production with division of labour in manu- 
factories and factories comes into life, and with it the 
wage-earning class. In place of- the mediaeval exchange 
bank there is developed first the bank of deposit and cir- 
culation, then the modern credit bank. The transport of 
goods, which earlier was an integral part of trade, now 
becomes independent. The state posts, newspapers, and 
the national marine arise, and the insurance system is de- 
veloped. On all sides are new organizations whose pur- 
pose is to satisfy wide-spread economic wants; a national 
industry, a national market, national commercial institu- 
tions,— everywhere the capitalistic principle of business 
enterprise in trade. 

Everybody knows how the absolutist State furthered 
this movement, and how not infrequently in the effort to 
accelerate it it gave an artificial existence to what would 
not flourish of its own strength. Nevertheless, though 
limited in manifold ways by state legislation, the old eco- 
nomic organization of the towns with its guild and monop- 
oly privileges and the sharp separation of town and 
country, persisted on until about the end of the eighteenth 



THE RISE OF NATIONAL ECONOMY. 139 

century, heedless of the new economic life springing up 
roundabout and of the variety of new commercial forms 
which it had nurtured. When the Physiocrats and Adam 
Smith made these latter for the first time the subject of 
scientific observation, they entirely overlooked the ob- 
vious fact that they were not dealing with a spontaneous 
product of mere social activity, but with a fruit of the 
paternal government of the State. The barriers whose 
removal they demanded were either the fossilized survivals 
of earlier economic epochs, such as charges upon the soil, 
guilds, local coercive rights, restrictions on freedom of mi- 
gration; or they were devices of mercantilism for assisting 
production, such as monopolies and privileges, which 
might cease to operate after having fulfilled their purpose. 

As far as the development of national economy is con- 
cerned, the liberalism of the last hundred years has only 
continued what absolutism began. Expressed in this way, 
the assertion may easily seem paradoxical. For liberalism, 
outwardly considered, has only demolished; it has over- 
thrown the antiquated forms ufKJn which household and 
town economy were founded, and constructed nothing 
new. It has. destroyed the special position and special 
privileges of individual territorial districts and individual 
social groups, and in their stead established free competi- 
tion and equality before the law. But though it has thus 
decomposed into its elements the heritage of the past, it 
has at the same time cleared the way for new economic 
combinations of a truly national character, and made it 
possible for every energy, according to the technical devel- 
opment of the time, to enter into the service of the whole 
at the point where it is of the greatest usefulness. 

If liberalism has made the progress of national economy 
absolutely contingent upon social freedom of action, and 
thus taken an attitude in many respects hostile to 



140 THE RISE OF NATIONAL ECONOMY. 

the State, it has nevertheless failed to prevent the modem 
State, as such, from pursuing the path chosen by it as early 
as the sixteenth century, and leading to an ever-closer 
union of all sections of the people and of the national ter- 
ritory for the accomplishment of the steadily expanding 
tasks of civilization. All the great statesmen of the last 
three centuries, from Cromwell and Colbert to Cavour and 
Bismarck, have worked towards this end. The French 
Revolution has been no less centralizing in its effects than 
the political upheavals of recent decades. In the latest 
phase of this evolution the principle of nationality has be- 
come a principle of mighty unifying power. The small 
separate States of earlier times were no longer equal to 
the comprehensive economic tasks of the present. They 
had either to disappear in one large national State, as in 
Italy, or surrender considerable portions of their inde- 
pendence, especially in economic legislation, to a federal 
State, as did the individual States in the German Empire, 
and the cantons in Switzerland. 

We err, if we imagine ourselves justified in concluding 
from the extent to which international trade has been 
facilitated during the epoch of liberalism that the period 
of national economy is on the decline and is giving place 
to the period of world-economy. The very latest political 
development of the States of Europe has resulted in a 
return to the ideas of mercantilism and, to a certain extent, 
of the old town economy. The revival of protective duties, 
the retention of national currency and of national labour 
legislation, the public ownership of the machinery of trans- 
portation already achieved or still aimed at, the national 
control of workman's insurance and of the banking sys- 
tem, the growing activity of the State in economic matters 
generally, — all this indicates that we have passed the ab- 
solutist and liberalist periods and entered upon a third 



*> 



THE RISE OF NATIONAL ECONOMY, 14 1 

period of national economy. Socially this period bears a 
peculiar aspect. It is no longer merely a question of meet- 
ing national wants as independently and completely as 
possible by national production, but a question of the just 
distribution of goods, of the direct action of the State in 
the economic interests of the whole, and for the puri>ose 
of securing to all its subjects according to their economic 
services a share in the benefits of civilization. The requi- 
site measures can be carried out only on a grand scale; 
they demand an intimate union of all individual forces, 
such as a great national State alone can furnish. 

With this we might fittingly close. For to present here 
the multitude of new phenomena springing up under the 
touch of national as opposed to household and town econ- 
omy one would need to reproduce almost the whole con- 
tents of a text-book on political economy. It will never- 
theless contribute to a better understanding of the subject 
if, by a comparison of some of the leading phenomena we 
concisely review the fundamental features of the whole in 
the three stages of its development. 

The most prominent of these features is, that in the 
course of history mankind sets before itself ever higher 
economic aims and finds the means of attaining these in a 
division of the burden of labour, which constantly extends 
until finally it embraces the whole people and requires the 
services of all for all. This cooperation is based, in the 
case of household economy, upon blood-relationship, of 
town economy upon contiguity, and of national economy 
upon nationality. It is the road traversed by mankind in 
passing from clanship to society, which, as far as we 
can see, ends in an ever-tightening social organization. On 
this road the means for satisfying the wants of the indi- 
vidual continually grow in fulness and variety, and at the 
same time in dependence and complexity. The life and 



142 THE RISE OF NATIONAL ECONOMY, 

labour of every individual becomes more and more en- 
twined with the life and labour of many others. 

At the stage of household economy every commodity is 
consumed in the place of its origin; at the stage of town 
economy it passes immediately from the producer to the 
consumer; at the stage of national economy, both in its 
production and thereafter, it passes through various hands 
— ^it circulates. In the course of the whole evolution the 
distance between production and consumption increases. 
At the first stage all commodities are consumption goods; 
at the second part of them become articles of exchange; 
at the third most of them are wares. 

The individual household at the first stage is a producing 
and consuming community in one; at the stage of town 
economy this state of things continues in so far as the 
journeyman craftsman and the peasant workman make 
part of the household of the person employing them; in 
national economy community in production and com- 
munity in consumption become distinct. The former is a 
business undertaking from whose returns as a rule several 
independent households are supported. 

When outside labour is necessary, it is at the first stage 
in a permanent relation of subjection to the producer (as 
slaves and serfs), at the second in one of service, and at the 
third the relationship is contractual. Under the inde- 
pendent household system the consumer is either himself 
a labourer, or the owner of the labourer; in town economy 
he makes a direct purchase of the workman's labour 
(wage- work), or of the product of his labour (handicraft); 
in national economy he ceases to stand in any relation to 
the labourer, and purchases his goods from the entrepre- 
neur or merchant, by whom the labourer is paid. 

As for money, it is in independent domestic economy 
either entirely absent, or an article of direct use and a 



THE RISE OF NATIONAL ECONOMY. 143 

means for storing up wealth. In town economy it is es- 
sentially a medium of exchange; in national economy it 
becomes a means of circulation and of profit-making as 
well. The three categories, payment in kind, money pay- 
ment, and payment based upon credit, correspond with the 
various roles played by money, though they do not ex- 
haust them. 

Capital scarcely exists at the first stage; we meet only 
with consumption goods. At the second stage implements 
of labour may be classed under the usual head of business 
capital, but this is by no means generally true of the raw 
materials. Acquisitive capital proper exists only in the 
form of trade capital. At the third, acquisitive capital rep- 
resents the means whereby goods are raised from one 
stage of division of labour to the next and impelled 
through the whole process of circulation.^® Here every- 
thing becomes capital. From this point of view we might 
describe the indepcHdent household economy as lacking 
capital, town economy as hostile to capital, and national 
economy as capitalistic. 

Income and wealth under the household system com- < 
pose an undivided and indivisible whole; though the be- 
ginnings of ground-rent are already perceptible. In town 
economy interest also usually appears as ground-rent; 
business profits are confined almost entirely to trade; the 
chief form of labour-wage is the wage paid to the crafts- 
man by the consumer. But even yet most commodities 
do not pass from their place of production into other es- 
tablishments. Pure income can be realized only by one 
who definitely surrenders a portion of his wealth in the 
purchase of rents. At the stage of national economy the 
four branches of income are definitely separated. Almost 

* Comp. also Chaps. IV and VIII. 



144 THE RISE OF NATIONAL ECONOMY. 

the whole return from production is liquidated through 
trade. Under wealth, rent and acquisitive capital are dis- 
tinct from stores for consumption, which are kept at the 
lowest imaginable limit, since commerce frees individuals 
from the keeping of supplies. On the other hand the un- 
used surpluses of income, which at the first and second 
stages necessarily remained over from the wealth available 
for consumption, are now either directly added to business 
capital or transformed by means of saving- and other 
banks into interest-bearing loans, — that is, they are, in 
any case, converted into capital. 

At the stage of household economy the division of labour 
is confined to the household establishment; at the stage of 
town economy it consists either in the formation of, and 
division into, trades within the town, or in a partition of 
production between town and country; while the promi- 
nent features of the stage of national economy are increas- 
ing division of production, subdivisioH of work within the 
various establishments, and displacement of labour from 
one business to another.^^ 

Industry as an independent occupation is not found at 
the first stage, the whole transformation of raw material 
being merely housework. In the town economy we indeed 
find labourers pursuing some special industrial occupa- 
tion, but entrepreneurs are lacking; industry is either, 
wage-work or handicraft, and he who wishes to ply it must 
first master it. In national economy industry carried on in 
factories and under the commission system is preponder- 
ant; and this presupposes extensive capital and an entre- 
preneur with mercantile skill. Technical mastery of the 
process of production by the entrepreneur is not indis- 
pensable.^® 

" For further details see Chap. VIII. 
" Comp. Chap IV. 



THE RISE OF NATIONAL ECONOMY. 145 

In similar fashion a change occurs in the forms under 
which trade is pursued. Corresponding to the household 
system is itinerant trade, to town economy market trade, 
and to national economy trade with permanent establish- 
ment. If at the first two stages trade is merely supple- 
mentary to an otherwise autonomous system of produc- 
tion, it becomes in national economy a necessary link 
between production and consumption. It draws away 
from transportation, which now attains an independent 
position and organization. 

Commercial services were, to be sure, not lacking in the 
ancient slave and the mediaeval manorial systems; they de- 
volved upon special slaves or serfs. In the Middle Ages 
we find town messengers who were originally in the ex- 
clusive service of the municipal authorities, but later added 
the carriage of private correspondence. At the threshold 
of modem times stands the postal service, at first restricted 
to state purposes, by-and-bye extended to the public. In 
our century follow the railway, telegraph, telephone, and 
steamship lines — ^with which the State interferes in the 
interest of economy — and along with them the most varied 
private undertakings for facilitating communication.^* 
At-all the stages, however, certain commercial services have 
been organized by the sovereign administration, in the 
initial instance always for its own special requirements. 

Credit is at the first stage purely consumption credit; 
and can be obtained only by the person pledging himself 
and all his property. At the second stage, in the matter of 
personal credit, servitude for debt is softened to imprison- 
ment for debt. Along with consumption credit appears a 
type of credit on the return from immovables which is 
met in garb of a purchase, and must be considered as the 

■• For the analogous development in the newspaper press see Chap. 
VI. 



14^ THE RISE OF NATIONAL ECONOMY. 

normal form of credit under town economy. Business or 
productive credit, the distinctive form of credit in modern 
times, is first developed in trade, whence it spreads to 
every sphere of industrial life. State credit appears in the 
States of antiquity naturally as a forced loan; in the medi- 
aeval towns as the sale of annuities and redeemable claims; 
in the modern States as the disposing of perpetual rents 
or of redeemable interest-bearing bonds. 

In the domain of public services similar stages may also 
be pointed out. Legal protection is at first a matter for 
the clan, later for the feudal lord; in the Middle Ages the 
towns form districts of separate jurisdiction; at present 
the enforcement of law and police protection are functions 
of the State. The same is the case with education. At 
the first stage education devolves upon the family, as it does 
to-day still in Iceland. The Roman pcedagogus is a slave. 
In the Middle Ages it is autonomous household establish- 
ments, namely, the monasteries, that organize the educa- 
tional system; later arise the municipal and cathedral 
schools; peculiar to modem times are the concentration 
and supervision of instruction in state institutions. This 
development is even more apparent in the arrangements for 
defence. Among many peoples still at the stage of eco- 
nomic isolation each separate house is fortified (for ex- 
ample, the palisades of the Malays and Polynesians), and 
in early mediaeval times the manor is protected by wall and 
moat. At the second economic stage each city is a fort- 
ress; at the third a few fortifications along the borders 
secure the whole State. It is sufficiently significant that 
Louvois, the creator of the first system of border fortifica- 
tion, was a contemporary of Colbert, the founder of mod- 
ern French national economy. 

These parallels might be multiplied. As in moving into 
a new building one's first care is to introduce order for 



THE RISE OF NATIONAL ECONOMY. I47 

the time being, so with regard to the subject-matter of this 
chapter no fair-minded person will expect that everything 
has been exhaustively treated and every detail assigned its 
proper place. The writer clearly perceives how inade- 
quately the various phenomena of the two earlier stages 
of industrial evolution have as yet been worked over, and 
how very seriously their economic significance still de- 
mands accurate ascertainment. But for the present it may 
suffice if we have made clear the regularity of development 
both generally and in detail. 

Only one thing further would we particularly em- 
phasize. Household economy, town economy, national 
economy — these phrases do not denote a series whose 
terms arc mutually exclusive. One kind of economic life 
has always been the predominant, and in the eyes of con- 
temporaries the normal, one. Many elements of town 
economy, and even of independent domestic economy, 
still project into the present. Even to-day a very consid- 
erable part of the national production does not pass into 
general circulation, but is consumed in the households 
where it is produced; another portion, again, circulates 
no farther than from one establishment to another. 

Hence it would almost appear as if those people were 
in error who regard the task of political economy to be 
the explanation of the nature and coherence of commercial 
phenomena, and those right who confine themselves to a 
description of economic forms and their historical trans- 
formations. 

Yet that would be a fatal error, involving the sur- 
render of the scientific labours of over a century, as well 
as a complete misconception of our economic present. 
To-day not a sack of wheat is produced even on the most 
remote farm, that is not directly linked to the industrial 
life of the nation as a whole. Even if it be consumed in 



148 THE RISE OF NATIONAL ECONOMY. 

the house of the producer, nevertheless a large portion of 
the means of its production (the plough, the scythe, the 
threshing-machine, the artificial fertilizers, the draught- 
animals, etc.) is obtained through trade; and the con- 
sumption of one's own products takes place only when 
from market conditions it seems economically advisable. 
Thus the sack of wheat is knit by a strong cord to the great 
intricate web of. national commerce. And so are we all in 
our every economic thought and deed. 

It is therefore a matter of great satisfaction that, after 
a period of diligent collection of material, the economic 
problems of modern commerce have in recent times been 
zealously taken up again, and that an attempt is being 
made to correct and develop the old system in the same 
way in which it arose, with the aid, however, of a much 
larger store of facts. For the only method of investigation 
which will enable us to approach the complex causes of 
commercial phenomena is that of abstract isolation, and 
logical deduction. The sole inductive process that can 
likewise be considered, namely, the statistical, is not suffi- 
ciently exact and penetrating for most of the problems 
that have to be "handled here, and can be employed only 
to supplement or control. 

For the economic periods of the past the task will not be 
different. Here, to be sure, it will be even more necessarj' 
first to collect the facts and present them according to 
form and function; next we must gain a proper conception 
of the nature of the phenomena; and then we may logically 
dissect them and investigate their casiial connection. We 
will thus have to advance by the same method that " classi- 
cal political economy " has applied to the industry of the 
present. For some phases of the economic life of the an- 
cient ohco^ this has already been done in a masterful man- 
ner by Rodbertus; for the economic life of the Middle 



THE RISE OF NATIONAL ECONOMY. 149 

Ages such a task has yet scarcely been essayed. The at- 
tempt can succeed only with investigators fully able to 
grasp the actual assumptions of past economic periods 
and our ancestors* ways of thinking on economic matters; 
it can but fail if the half understood, half arbitrarily recon- 
structed economic conditions of the past continue to be 
reflected in terms of the modem theory of exchange. 

Only in this way, in our opinion, can investigations in 
economic history and the theory of contemporary eco- 
nomics be mutually helpful; only thus can we gain a 
clearer insight into the regularity both of economic devel- 
opment and of economic phenomena. 



CHAPTER IV. 

A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS, 

In economic and social matters most people have very 
definite opinions on what should be, often much more 
definite than on what is. What in their view should be is 
by no means an ideal state of affairs, an imaginative crea- 
tion that has never been realized. Very frequently indeed 
it is a conception drawn from the conditions that prevailed 
in times more or less remote, which long custom has led 
us to consider normal. 

Such is the case, if we mistake not, with many of our 
contemporaries regarding what we call handicraft and the 
so-called handicraft problem. One has become accus- 
tomed to look upon handicraft as the normal form of in- 
dustry, after it has dominated five centuries or more of 
the life of the burgher class of Germany. The proverb 
says " Handicraft stands on golden ground "; and obser- 
vation teaches us that this ground is, according to present- 
day valuation, no longer golden. We ask ourselves how 
that happy condition can be restored, how handicraft can 
be " resuscitated." 

But what right has one to regard handicraft as the 
normal form of industry and thus as it were to strive after 
an ideal whose realization belongs to the past? 

The earlier political economists represent handicraft as 

the original form of industrial production. " In a tribe of 

150 



A HISTORICAL SURREY OF INDUSTRIAL SYSTEMS. 151 

hunters or shepherds," says Adam Smith/ " a particular 
person makes bows and arrows with more readiness and 
dexterity than any other. He frequently exchanges them 
for cattle or venison with his companions; and he finds at 
last that he can in this manner get more cattle and venison 
than if he himself went to the field to catch them." 
Finally, ** the making of bows and arrows grows to be his 
chief business and he becomes a sort of armourer." If we 
follow this historical progress a couple of stages further, 
the original handicraftsman will after a time probably take 
an apprentice, and when the latter has learned his trade, a 
second, while the first becomes his journeyman. 

Seek as we may, we find nothing added by subsequent 
development. When we speak of a craftsman to-day we 
have in mind a business undertaker on a small scale, who 
has passed by regular stages of transition from apprentice 
to journeyman and from journeyman to master workman, 
who produces with his own hand and his own capital for 
a locally limited circle of customers, and into whose hands 
flows undiminished the whole product of his labour. 
Everything that one can demand of an industrial system 
founded on justice seems realized in the life of the typical 
craftsman — gradual social progress, independence, an in- 
come corresponding to services rendered. And those 
forms of industry that vary from this primal type, namely, 
house industry and factory production, may readily ap- 
pear abnormal; and the social stratification of those em- 
ployed, and the accompanying unequal distribution of 
income out of harmony with the idea of economic justice. 

Even later economists are rarely free from this popular 
conception. In contrasting the three industrial systems 
that they recognise, handicraft, house industry, and factory 

* Bk. I, ch. 2. 



152 A HISTORICAL SURREY OF INDUSTRIAL SYSTEMS. 

production, they almost unwittingly draw from the funda- 
mental institutions of handicraft the criterion for judging 
the others. Until quite recently house industry was for 
many of them merely a degenerate handicraft or a transi- 
tional form, and the factory a necessary evil of the age of 
machinery. This narrowness of view was prejudicial to the 
scientific understanding of even modern industrial meth- 
ods, open as these are to direct observation. 

An historically constructive view, such as we will here 
present, must from the start shake off the idea that any 
particular form in any department of economic activity 
can be the norm for all times and peoples. Even handi- 
craft is for it only one phenomenon in the great stream 
of history, with its origin, continuance, and success de- 
pendent upon certain given economic conditions. It is 
neither the original nor even a necessary form in the his- 
torical evolution of industrial production. It is, in other 
words, just as little necessary that the industry of a country 
shall have passed through the handicraft phase before ar- 
riving at house industry or factory manufacture as that 
every people shall have been hunters or nomads before 
passing over to settled agriculture. Among us handicraft 
has been preceded by other industrial systems, which, in- 
deed, even in Europe, still exist in part. 

The great historical significance of these primitive in- 
dustrial forms in the evolution of economic conditions has 
hitherto been almost wholly ignored, although they shaped 
for thousands of years the economic life of the nations and 
left lasting marks upon their social organizations. Only a 
comparatively small portion of the history of industry, 
namely, that part which written laws have enabled us to 
know, has been at all cleared up; and this, too, much more 
on its formal side than as regards its inner life, its method 
of operation. Even the guild handicraft of the Middle 



A HISTORICAL SURREY OF INDUSTRIAL SYSTEMS. 153 

Ages, to which in recent times so much persevering and 
penetrating labour has been devoted, has, on the side of 
its actual operation, enjoyed scarcely more accurate in- 
vestigation. In this domain arbitrary theoretical con- 
structions based upon the postulates and concepts of 
modern commercial economy still widely prevail. 

Our " historical " political enonomy, it is true, has a 
wealth of material for the economic history of the classical 
and modem peoples. But it has hardly yet been duly 
noted that the complex nature of all social phenomena ren- 
ders it just as difficult for thp investigator of to-day to re- 
construct the economic conditions of the life of the na- 
tions of antiquity and of the Middle Ages as to forecast 
even with the most lively and powerful imagination the 
ultimate consequences of the " socialist State of the fu- 
ture." We shall not arrive at an understanding of whole 
epochs of early economic history until we study the eco- 
nomic side of the life of primitive and uncivilized peoples 
of the present with the care we to-day devote to English- 
men and Americans. Instead of sending our young politi- 
cal economists on journeys of investigation to these latter, 
we should rather send them to the Russians, the Rumani- 
ans, or the South Slavs; we should study the characteristic 
features of primitive economic life and the legal concep- 
tions of the peoples of our newly acquired colonies before 
such features and conceptions disappear under the influ- 
ence of European trade. 

It is almost a fortunate circumstance that such external 
influences rarely affect deeply the real life of the people, 
but are confined chiefly to the more privileged classes. 
Hence it is that in extensive regions of eastern and north- 
ern Europe, which the unheeding traveller courses 
through by rail, there may still be observed among the 



154 A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. 

rural population primitive forms of production that mod- 
em commerce has caused to vary but slightly. 

In the attempt made in the following pages to give a 
compact presentation of what we know of the industrial 
methods of such " backward " tribes and the present con- 
clusions of industrial history, our sole aim is to present 
in clear outline the chief stages of development.^ In or- 
der to have a guiding thread through the perplexing va- 
riety and wealth of forms of individual ethnographical 
observations, it is most necessary to separate typical and 
casual, to disregard subsidiary and transitional forms, and 
to consider a new phase of development as beginning only 
where changes in industrial technique call forth economic 
phenomena that imply a radical alteration in the organ- 
ization of society. In this way we arrive at five main sys- 
tems of industry. In historical succession they are: 

1. Housework (Domestic Work). 

2. Wage-work. 

3. Handicraft. 

4. Commission Work (House Industry). 

5. Factory Work. 

We shall first attempt to give a concise outline of 
the characteristic economic peculiarities of these indus- 
trial systems, merely indicating the socio-historical im- 
port of the whole development. The filling out of oc- 
casional gaps and the explanation of the transitions from 
one system to the other may be left to detailed investiga- 
tion. In our sketch we shall, naturally, devote most time 
to the two industrial systems precedent to handicraft, 

* The present sketch brings together in popular form only the most 
important features. Further discussion, and references to the most 
essential literature, will be found in my article under the head " Gc- 
werbc" in the Hand wort, der Staatswiss. (2d ed.), IV, pp. 360-393. 



A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. 155 

while for the later a brief account may suffice. We begin 
with housework.® 

Housework is industrial production in and for the house 
from raw materials furnished by the household itself. In 
its original and purest form it presupposes the absence 
of exchange, and the ability of each household to satisfy 
by its own labour the wants of its members. Each com- 
modity passes through all the stages of production in the 
establishment in which it is to be consumed. Production 
is consequently undertaken only according to the needs of 
the house itself. There is still neither circulation of goods 
nor capital. The wealth of the house consists entirely in 
consumption goods in various stages of completion, such 
as corn, meal, bread, flax, yarn, cloth, and clothes. It also 
possesses auxiliary means of production, such as the hand- 
mill, the axe, the distaff, and the weaver's loom, but no 
goods with which it could procure other goods by pro- 
cess of exchange. All it has it owes to its own labour, 

•It is from Norway and Sweden that the expression HausAeisz 
(housework) has been transplanted into Germany, where during the 
last twenty years it has become current. In those countries it was 
employed for certain occupations of the members of the domestic 
circle, such as spinning, weaving, sewing, the making of wooden 
utensils, and the like. It is the application of industrial technique 
that, favoured by climate and settlement, has from early times become 
indigenous in those parts. Through it the peasant household works 
up for its own use the raw material from field and wood. As this 
technique threatened to disappear under the influence of modem com- 
mercial Conditions it was the opinion in Denmark and Norway that 
it should be reanimated through school instruction. Few, indeed, of 
the promoters of manual instruction — ^this new branch of instruction, 
whose pedagogical importance cannot be denied — have formed a clear 
conception of what housework really signified and in part still signifies 
for the Northern peoples. Here and there, especially at its inaugura- 
tion, manual training was regarded as a means for establishing new 
house industries. But housework and house industry are two indus- 
trial systems separated historically from one another (at least among 
us) by two centuries. 



156 A HISTORICAL SURREY OF INDUSTRIAL SYSTEMS. 

and it is scarcely possible to separate the operations of the 
household from those of production. 

In the form of housework, industry is older than agri- 
culture. Wherever explorers of new countries have come 
into contact with primitive peoples, they have found many 
forms of industrial skill, such as the making of bow and 
arrow, the weaving of mats and vessels out of reeds, bast, 
and tough roots, a primitive pottery, tanning skins, crush- 
ing farinaceous grains on the* grinding-stone, smelting 
iron ore, the building of houses. To-day the hunting 
tribes of North America, the fisher tribes of the South Sea, 
the nomad hordes of Siberia, and the agricultural negro 
tribes of Africa make similar display of varied technical 
skill without possessing actual artisans. Even the wretched 
naked forest tribes of Central Brazil make their clubs and 
bows and arrows, build houses and bark canoes, make 
tools of bone and stone, weave baskets for carrying and 
storing, scoop out gourd dishes, spin, knit, and weave, 
form artistically ornamented clay vessels without a knowl- 
edge of the potter's wheel, carve ornamented digging- 
sticks, stools, flutes, combs, and masks, and prepare many 
kinds of ornaments out of feathers, skins, etc. 

In the temperate and colder countries with the advance 
to the use of the plough, this activity loses more and more 
the character of the accidental; the whole husbandry ac- 
quires a settled character; the mild period of the year must 
be devoted to the procuring of raw material and to out- 
door work; in winter the working up of this material 
clusters the members of the household around the hearth. 
For each kind of work there is developed a definite method 
which is incorporated into the domestic life according to 
the natural and imperative demands of economy; about it 
custom weaves its fine golden ethical thread; it enriches 
and ennobles the life of men among whom, with its sim- 



A HISTORICAL SURREY OF INDUSTRIAL SYSTEMS. 157 

pie technique and archaic forms, it is transmitted from gen- 
eration to generation. As people labour only for their 
own requirements, the interest of the producer in the work 
of his hands long survives the completion of the work. His 
highest technical skill and his whole artistic sense are em- 
bodied in it. It is for this reason that the products of do- 
mestic work throughout Germany have become for our 
age of artistic industry such a rich mine of models of pop- 
ular style. 

The Norwegian peasant is not merely his own smith and 
joiner, like the Westphalian Hofschulze in Immermann's 
" Miinchhausen "; with his own hands he also builds his 
wooden house, makes his field-implements, wagons and 
sleighs, tans leather, carves from wood various kinds of 
house utensils, and even makes metal ones.*^ In Iceland 
the very peasants® are skilful workers in silver. In the 
Highlands of Scotland, up to the close of last century, 
every man was his own weaver, fuller, tanner, and shoe- 
maker. In Galicia and Bukowina, in many parts of Hun- 
gary and Siebenbiirgen, in Rumania, and among the 
southern Slav peoples there could scarcely be found, down 
to recent times, any other craftsman than the smith, and 
he was usually a gypsy. In Greece and other lands of the 
Balkan peninsula the only additional craftsmen were oc- 
casional wandering builders.'^ Numberless examples of a 

• Eilert Sundt, Dm HusHiden i Norge (Christiania, 1867). Blom, Das 
Konigreich Norwegen (Leipzig, 1843), p. 237. Forester, Norway in 
1848 and 1849 (London, 1850), p. 113. E. Sidenbladh, Schweden, Statist, 
Mitteilungen g, Wiener Weltausstellung, 1873. 

• [Dr. Biicher is evidently speaking of the Icelandic peasant of an 
earlier time. — Ed.] 

• On the Austrian populations compare Die Hawindustrie Oesterreichs. 
Ein Kommentar z. hausindustriellen Ahteilung auf d. aUgemeinen land- «. 
forstwirtsehaftlicK Ausstellung zu Wien, 1890, ed. by W. Exner; also 
Oesterreichische Monatsschrift fiir Gesellschaftswissenschaft, IV, 
goff., VIII, 22, IX, 98 and 331; A. Riegl, Textile Hausindustrie in 



15^ A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. 

similar kind might be adduced from other peoples. The 
wonderful adroitness and dexterity of the Russian and 
Swedish peasants, to cite a striking instance, has its un- 
doubted origin in the varied technical tasks of their own 
households. The industrial employments of women in 
ancient and modern times, such as spinning, weaving, bak- 
ing, etc., are too well known to call for further reference. 

In order to obtain an idea of the wealth of domestic in- 
dustrial skill that characterizes the life of less civilized peo- 
ples a detailed description would be necessary. Lack of 
space unfortunately forbids that here. It will suffice, how- 
ever, to reproduce the following sentences from an account 
of household work in Bukowina:® 

" In the narrow circle of the family, or at least within 
the limits of his little village, the Bukowina countryman 
supplies all his own necessaries. In building a house the 
husband, as a rule, can do the work of carpenter, roofer, 
etc., while the wife must attend to plastering the woven 
and slatted walls or stopping the chinks in the log walls 
with moss, pounding out of the floor, and many other 
related duties. From the cultivation of the plant from 
which cloth is spun or the raising of sheep down to the 

Oesterreich in the Mitteilungen d, k. k. osterreich. Museums, New 
Series, IV, pp. 411 ff.; Braun und Krejcsi, Der HausHeiss in Ungam 
(Leipzig, 1886) ; Schwicker, Statistik d, Konigreichs Ungarn^ pp. 403 ff., 
411, 426 ff.; J. Paget, Ungam u. Siehenburgen (Leipzig, 1842), II, pp. 
163, 173, 264, 269; Franz Joseph Prinz von Battenberg, Die Volks- 
wirtschaftl. Entwickelung Bulgariens (Leipzig, 1891); Iwantschoff, 
Primitive Formen d. Gewerbebetriebs in Bulgarien (Leipzig, 1896). On 
the other lands of the Balkan peninsula see Reports from Her Majesty's 
Diplomatic and Cottsular Agents Abroad, respecting the Condition of the 
Industrial Classes in Foreign Countries (London, 1870-72); Tarajanz, 
Das Gewerbe bei d. Armeniern (Leipzig, 1897); Petri, Ehstlaftd u, d. 
Ehsten (Gotha, 1802), II, pp. 230-1. 

•C. A. Romstorfer in Exner, Die Hausindustrie Oesterreichs, pp. 
159 ff. Comp. H. Wiglitzky, Die Bukowinaer Hausindustrie u, d. Mittel 
u, Wege s, Hebung derselb. (Czernowitz, 1888). 



A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. 159 

making of bed and other clothes out of linen, wool or furs, 
leather, felt, or plaited straw, the Bukowina country folk 
produce everything, including dyes from plants of their 
own culture, as well as the necessary though, indeed, ex- 
tremely primitive utensils. The same holds in general 
of the food-supply. With a rather heavy expenditure of 
labour the peasant cultivates his field of maize, and with 
his handmill grinds the kukuruz m^eal used by him in bak- 
ing mamaliga, his chief article of food, which resembles 
polenta. His simple farming implements, the dishes and 
utensils for household and kitchen, he, or, if not he, some 
self-taught villager, is also able to make. The working of 
iron, alone, a substance that the native population uses in 
exceedingly small quantities, he generally leaves to the 
gypsies scattered through the country.*' 

Yet whatever the industrial skill developed by the self- 
sufficing household, such a method of supply was destined 
to prove inadequate when the household diminished to the 
smaller circle of blood-relations, which we call the family. 
The ancient family group, it is true, was broader than our 
present family; but just at the time when wants are in- 
creasing in extent and variety, the tribal organization of 
many peoples breaks down and a more minute division of 
labour among the members of the household is rendered 
impossible. The transition to specialized production and 
a system of exchange would at this point have been un- 
avoidable had it not been possible, by adopting slaves or 
by utilizing serf labour, to enlarge artificially the house- 
hold circle. The greater the number of these unfree mem- 
bers of the household, the easier it is to introduce a varied 
division of labour among them and to train each person 
for a definite industrial employment. 

Thus we find among the house-slaves of the wealthy 



l6o A HISTORICAL SURREY OF INDUSTRIAL SYSTEMS. 

Greeks and Romans industrial workers of various kinds; • 
and in the famous instructions of Charles the Great regard- 
ing the management of his country estates we have definite 
rules prescribing what kinds of unfree workers shall be 
maintained at each villa. " Each steward," we read, 
" shall have in his service good workmen, such as smiths, 
workers in gold and silver, shoemakers, turners, car- 
penters, shield-makers, fishers, fowlers, soap-boilers, brew- 
ers of mead (siceratores), bakers, and net-makers." Co- 
pious evidence of a similar kind is available for the manors 
of the nobility and the monasteries. The handicraftsmen 
maintained by them are at their exclusive service; in some 
cases they are merely domestic servants receiving their 
board and lodging in the manor-house, in others they are 
settled and gain their living on their own holdings, and in 
return render villein services in that branch of labour in 
which they have special skill. In token that they are en- 
gaged to hold their skill at the service of the manor, they 
bear the title oiHciales, ofRciati, i.e., officials. 

Housework, we see, has here obtained an extensive or- 
ganization, which allows the lord of the manor a relatively 
large and varied consumption of industrial products. 

But housework does not remain mere production for 
direct consumption. At a very early stage inequality of 
natural endowment causes a varied development of techni- 
cal skill. One tribe produces pottery, stone implements, or 
arrows, and a neighbouring tribe does not. Such industrial 
products are then scattered among other tribes as gifts 
of hospitality, or as spoils of war, and later as the objects 
of exchange.® Among the ancient Greeks wealthy slave- 
owners caused a considerable number of their dependent 

• Comp. H. Francotte, LIndustrie dans la Grhe aticienne, I (Brussels, 
1900) ; Wallon, Hist de FEsclavage dans TAnHquiti (2d ed., Paris, 1879). 

• Comp. above, pp. 54 flF.* 



A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. i6i 

labourers, whom they did not need for their own estates, to 
be trained for a special industry, and then to produce for 
the market. In a similar fashion peasant families exchange 
the surplus products of their household industry more fre- 
quently than the surpluses from their agriculture or cattle- 
raising. As in the Old Testament it is one of the good 
qualities of the virtuous wife to dispose of the wares that 
her own hands have produced,® so to-day the negro wife in 
Central Africa carries to the weekly market the pots or 
basketware she produces in order to exchange them for 
salt or pearls. In like manner, in many parts of Germany 
the rural population have from the beginning of the 
Middle Ages sold their linen cloths at the town mar- 
kets and fairs; and in the era of mercantilism measures 
were taken by the government in Silesia and West- 
phalia to facilitate the export of home-made linen. So 
also in the Baltic provinces during the Middle Ages the 
coarse woollen cloth, Vadhtnal, which is still woven by the 
peasant women, was one of the best known articles of 
trade, and actually served as money. Similarly among 
many African peoples domestic products made by neigh- 
bouring tribes serve as general mediums of exchange. In 
almost every villager's house in Japan yarn is spun and 
cloth woven out of cotton grown in his own fields, and of 
this a portion comes into exchange. In Sweden the West 
Goths and Smalanders wander through almost the whole 
country offering for sale home-woven stuffs. In Hungary, 
Galicia, Rumania, and the southern Slav countries, every- 
where one can meet with peasants offering for sale at the 
weekly town markets their earthen and wooden wares, and 
peasant women selling, along with vegetables and eggs, 
aprons, embroidered ribbons, and laces which they them- 
selves have made. 

• F. Buhl, Die sozial. Verhdltnisse d. Israeliten (Leipzig, ^898), p. 34. 



l62 A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS, 

It is especially when the land owned by a family be- 
comes divided up and no longer suffices for its mainte- 
nance, that a part of the rural population take up a special 
branch of housework and produce for the market in ex- 
actly the same way as our small peasants in South Ger- 
many produce wine, hops, or tobacco. At first the neces- 
sary raw material is gained from their own land or 
drawn from the communal forests; later on, if need be, it 
is also purchased. All sorts of allied branches of produc- 
tion are added; and thus there develops out of housework, 
as in many parts of Russia, an endlessly varied system of 
peasant industry on a small scale. 

But the evolution may take another course, and an in- 
dependent professional class of industrial labourers arise, 
and with them our second industrial system — wage-work. 
Whereas all industrial skill has hitherto been exercised in 
close association with property in land and tillage, the 
adept house-labourer now frees himself from this associa- 
tion, and upon his technical skill founds for himself an ex- 
istence that gradually becomes independent of property 
in land. But he has only his simple tools for work; he has 
no business capital. He therefore always exercises his 
skill upon raw material furnished him by the producer of 
the raw material, who is at the same time the consumer of 
the finished product. 

Here again two distinct forms of this relationship 
are possible. In one case the wage-worker is taken tem- 
porarily into the house, receives his board and, if he does 
not belong to the place, his lodging as well, together with 
the daily wage; and leaves when the needs of his customer 
are satisfied. In South Germany we call this going on 
one's itinerancy {(xuf die Stor gehen), and may accordingly 
designate the whole industrial phase as that of itinerancy 
(Stor), and the labourer carrying on work in this manner 



A HISTORICAL SURREY OF INDUSTRIAL SYSTEMS. 1 63 

as the itinerant (Storer). The dressmakers and seam- 
stresses whom our women in many places are accustomed 
to take into the house may serve as an illustration. 

On the other hand, the wage-worker may have his own 
place of business, and the raw material be given out to him. 
For working it up he receives piece-^ork wage. In the 
country the linen-weaver, the miller, and the baker work- 
ing for a wage are examples. We will designate this form 
of work home work.^ It is met with chiefly in industries 
that demand permanent means of production difficult to 
transport, such as mills, ovens, weavers' looms, forges, etc. 

Both forms of wage-work are still very common in all 
parts of the world. Examples might be drawn from In- 
dia and Japan, from Morocco and the Sudan, and from 
almost all European countries. The system can be traced 
in Babylonian temple records and in ancient Egypt; it can 
be followed in literature from Homer down through an- 
cient and mediaeval times to the present day. The whole 
conception of the relation of the customer to the independ- 
ent (personally free or unfree) artisan in early Greek and 
Roman law rests upon wage-work;*^ and only by it are 
numerous ordinances of mediaeval guild law to be ex- 
plained. 

In the Alpine lands it is still the predominant industrial 
method in the country. The Styrian writer P. K. Roseg- 
ger has, in an interesting book, ^^ given a picture of his ex- 
periences as apprentice to a peripatetic tailor carrying on 
his trade among the peasants. " The peasant craftsmen," 

• Heimwerk. 

"In Diocletian's edict de pretiis rerum venalium of the year 301 it 
appears as the prevailing industrial form. Comp. my articles in the 
Ztschr, f, die ges. Staatswissenschaft, vol. 50 (1894), especially pp. 

673 ff . 
^ Aus meinem Handwerkerleben (Leipzig, 1880). Comp. also Hans- 

jakob, Schneeballen, First Series (Popular Ed.), pp. 12-13, 219-224. 

IVilde Kirschen, p. 347. 



l«)4 A HISTORICAL SURyEY OF INDUSTRIAL SYSTEMS. 

he says in the preface, " such as the cobbler, the tailor, the 
weaver, the cooper (in other places also the saddler, the 
wheelwright, the carpenter, and, in general, all artisan 
builders), are in many Alpine districts a sort of nomad folk. 
Each of them has, indeed, a definite abode somewhere, 
either in his own little house or in the rented room of a 
peasant's home, where his family lives, where he has safe- 
keeping for his possessions, and where he spends his Sun- 
days and holidays. On Monday morning, however, he puts 
his tools upon his back or in his pocket and starts out upon 
his rounds; that is, he goes out for work and takes up his 
quarters in the home of the peasant by whom he has been 
engaged, and there remains until he has satisfied the 
household needs. Then he wends his way to another farm. 
The handicraftsman in his temporary abode is looked upon 
as belonging to the family." Every peasant's house has a 
special room with a " handicraftsman's bed " for his 
quarters overnight; wherever he has been working during 
the week, he is invited to Sunday dinner. 

We find described in almost the same words the indus- 
trial conditions of rural Sweden and many parts of Nor- 
way. In Russia and the southern Slav countries there are 
hundreds of thousands of wage-workers, belonging espe- 
cially to the building and clothing trades, who lead a con- 
tinuous migratory life and who, on account of the great 
distances travelled, often remain away from home half a 
year or more. 

From the point of view of development these two forms 
of wage-work have different origins. Itinerant labour is 
based upon the exclusive possession of aptitude for a spe- 
cial kind of work, homework upon the exclusive posses- 
sion of fixed means of production. Upon this basis there 
now arises all sorts of mixed forms between housework 
and wage-work. 



A HISTORICAL SURyEY OF INDUSTRIAL SYSTEMS, 165 

The itinerant labourer is at first an experienced neigh- 
bour whose advice is sought in carrying out an important 
piece of work, the actual work, however, still being per- 
formed by the members of the household.^* Even later it 
is long the practice for the members of the customer's 
family to give the necessary assistance to the craftsman 
and his journeyman; and this is still met with in the 
country, for example, in the raising of a frame building. 

In the case of homework the later tradesman is at first 
merely the owner of the business plant and technical 
director of the production, the customer doing the actual 
work. This frequently remains true in the country to-day 
with oil-presses, flax-mills, mills for husking barley and 
oats, and cider-mills. 

In many North German towns the mediaeval maltsters 
and brewers were merely the owners of malt-kilns and 
brewing-houses, who for a fee gave the citizens the 
opportunity of malting their own barley and brewing their 
own beer. In the flour-mills the customer at least sup- 
plied the handler who attended to the sifting of the meal. 
Even to-day it is customary in many localities for the 
peasant's wife, after kneading the dough, to mould the 
bread-loaves in her own house; the baker simply places 
his oven at her disposal, heats it and attends to the bakinjg. 
In French and western Swiss towns the public washing 
places are managed in much the same fashion, merely pro- 

^* The same is true of house-building in the Caroline Islands, where 
the takelbay, or master builder, is scarcely more than the exerciser of 
the evil powers that threaten the new structure. See Kubary, Ethnogr, 
BeitrdgCf pp. 227 ff. The case is different with wagon-building in 
Armenia, where the skilled neighbour, in return for a present, directs 
the putting together of the vehicle after the separate parts have been 
made ready by the members of the household: Tarajanz, as above, 
p. 27. Similarly with house-building in Faror: Ztschr. d. Ver. L 
Volksk., Ill (1893), p. 163. 



l66 A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. 

viding their customers with washing-apparatus and hot 
water, and frequently a drying-place in addition, while the 
work is done by the servants or female members of the 
customer's household. These afterwards bring the washed 
and dried linen to the mangle to be smoothed out, in which 
process the owner assists by working the handle. Pay- 
ment is made by the hour. In Posen and West Prussia 
until recently it was the custom for the owner of a smithy 
merely to supply fire, tools, and iron, leaving the actual 
work to his customers.^* 

From the economic point of view the essential feature of 
the wage-work system is that there is no business capital. 
Neither the raw material nor the finished industrial 
product is for its producer ever a means of profit. 
The character and extent of the production are still deter- 
mined in every case by the owner of the soil, who pro- 
duces the raw material; he also superintends the whole 
process of production. The peasant grows, threshes, and 
cleans the rye and then turns it over to the miller to be 
ground, paying him in kind; the meal is given to the 
baker, who delivers, on receipt of a baker's wage and in- 
demnification for the firing, a certain number of loaves 
made from it. From the sowing of the seed until the mo- 
ment the bread is consumed the product has never been 



It 



Erlehnisse eines Geistlichen im ostl. Grenzgehiet, in the Tagl. Rund- 
schau, Unterh. Beilage, 1879, No. 258. A point of interest here is 
the supplying of the iron by the owner of the business, this method 
of carrying on work thus forming a transition to handicraft. There 
are also forms in which itinerancy and homework are mingled. To 
this class belongs the Russian migratory tailor, who in each vil- 
lage where he has customers rents a room for a time and does work 
for wages. So also, according to Tarajanz, the silversmiths in Ar- 
menia. In the latter country the owner of an oil-press has to provide 
his machine, the necessary workmen, and the oxen for driving it; the 
customer not only assists in the work himself, but he pays, and also 
boards, the workmen and supplies the fodder for the oxen. 



A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. 167 

capital, but always a mere article for use in course of prep- 
aration. No earnings of management and interest charges 
or middleman's profits attach to the finished product, but 
only wages for work done. 

Under certain social conditions, end where needs are 
very simple, this is a thoroughly economic method of pro- 
duction and, like housework, secures the excellence of the 
product and the complete adjustment of supply to demand. 
It avoids exchange, where this would lead only to a round- 
about method of supplying the producer of the raw mate- 
rial with wares prepared from his own products. But it 
also forces the consumer to run the risk attaching to indus- 
trial production, as only those needs that can be foreseen 
can find suitable and prompt satisfaction, while a sudden 
need must often remain unsatisfied because the wage- 
worker happens at the very time to be elsewhere engaged. 
In the case of homework there is the additional danger 
that a portion of the material furnished may be embezzled 
or changed. The system has also many disadvantages for 
the wage-worker. Amongst these are the inconveniences 
and loss of time suffered in his itinerancy from place to 
place; also the irregularity of employment, which leads 
now to the overwork, now to the complete idleness, of the 
workman. Both forms of wage-work thus act satisfac- 
torily only when the unoccupied hours can be turned to 
account in some allied branch of agriculture. 

In the Middle Ages, when this could be done, wage- 
work greatly facilitated the emancipation of the artisan 
from serfdom and feudal obligations, as it requires prac- 
tically no capital to start an independent business. It is a 
great mistake still common to look upon the class of 
guild handicraftsmen of the Middle Ages as a class of small 
capitalists. It was in essence rather an industrial labour- 
ing class, distinguished from the labourers of to-day by the 



1 68 A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. 

fact that each worked not for a single employer but for a 
large number of consumers. The supplying of the mate- 
rial by the customer is common to almost all mediaeval 
handicrafts; in many instances, indeed, it continues for 
centuries, even after the customer has ceased to produce 
the raw material himself and must buy it, as, for example, 
the leather for the shoemaker and the cloth for the tailor. 
The furnishing of the material by the master workman is a 
practice that takes slow root; at first it holds only for the 
poorer customers, but later for the wealthy as well. Thus 
arises handicraft in the sense in which it is generally under- 
stood to-day; .but alongside it wage-work maintains 
itself for a long time, even entering, in many cases, into 
the service of handicraft. Thus the tanner is wage-worker 
for the shoemaker and saddler, the miller for the baker, the 
wool-beater, the dyer, and the fuller wage-workers for the 
cloth-maker. 

In the towns itinerancy is the first of the two forms 
of wage-work to decline. This decline is considerably 
hastened by the interference of the guilds.^* The itiner- 
ancy was too suggestive of early villepage. In it the 
workman is, so to speak, only a special kind of day- 
labourer, who must temporarily become a subordinate 
member of another household. Consequently from the 
fourteenth century on we find the guild ordinances fre- 
quently prohibiting the master from working in private 
houses. To the same cause is to be ascribed the hatred 

" In this connection it may not be out of place to point out that, 
in the industrial limitation of those entitled to the privileges of the 
guild, the old housework was at the same time affected. In very many 
of the guild ordinances we find the regulation that the non-guildsman 
may do handicraftsman's work, but only in so far as the needs of his 
household demand, not for purposes of sale. The surplus house pro- 
duction for the market described above (pp. i6o, i6i) was thereby 
made impossible. 



^ 



A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS, 169 

displayed by the town craftsmen towards those of the 
country, because the migratory labour of the latter could 
not well be forbidden. Eventually itinerant or botcher ^^ 
becomes a general term of contempt for those who work 
without regular credentials from the guilds. In the North 
German towns the guild masters claimed the right of en- 
tering the hcmses of their customers to ferret out the itiner- 
ant artisans and call them to account, — the so-called 
•' botcherhunt "; and the public authorities were often 
weak enough to wink at this breach of the domestic rights 
of the citizen. 

But the guilds did not everywhere have such an easy 
task in supplanting one industrial system by another. As 
early as the middle of the fourteenth century the sovereign 
authority in the Austrian duchy takes vigorous measures 
against them. In the statutes of the electorate of Saxony 
for the year 1482 shoemakers, tailors, furriers, joiners, 
glaziers, and other handicraftsmen who shall refuse with- 
out sufficient reason to work in the house of their cus- 
tomer are made liable to a fine of three florins, a high sum 
for those times. In Basel a definite statute governing 
house tailors was enacted in 1526 for the maintenance of 
" ancient and honourable customs." In many German ter- 
ritories definite ordinances were made regulating the 
charges of the various kinds of wage-workers. Thus in 
many crafts, especially in the building trade, wage-work 
has persisted down to the present time. 

In the majority, however, its place has been taken by 
the industrial system that to-day is customarily desig- 
nated handicraft, whose nature we have indicated at the 
beginning of the presenf chapter. It might also be called 
price-work,^^ which would mark the contrast with wage- 

" Bonhasg. 
*• Preiswerk, 



lyo A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS, 

work. For the handicraftsman is distinguished from the 
wage-worker only by the fact that he possesses all the 
means of production, and sells for a definite price the fin- 
ished article which is the product of his own raw material 
and his own incorporated labour, while the wage-worker 
merely receives a recompense for his labour. 

All the important characteristics of handicraft may be 
summed up in the single expression custom production. It 
is the method of sale that distinguishes this industrial sys- 
tem from all later ones. The handicraftsman always works 
for the consumer of his product whether it be that the 
latter by placing separate orders affords the occasion for 
the work, or that the two meet at the weekly or yearly 
market. Ordered work and work for the market must 
supplement each other if " dull times " are to be avoided. 
As a rule the region of sale is local, namely, the town and 
its more immediate neighbourhood. The customer buys 
at first hand, the handicraftsman sells to the actual con- 
sumer. This assures a proper adjustment of supply and 
demand and introduces an ethical feature into the whole 
relationship; the producer in the presence of the con- 
sumer feels responsibility for his work. 

With the rise of handicraft a wide cleft, so to speak, 
appears in the economic process of production. Hitherto 
the owner of the land, though perhaps calling in the aid 
of other wage- workers, had conducted this whole process; 
now there are two classes of economic activity, each of 
which embraces only a part of the process of production, 
one producing the raw material, the other the manu- 
factured article. It is a principle that handicraft endeav- 
oured to carry out wherever possible — an article should 
pass through all the stages of its preparation in the same 
workshop. In this way the needed capital is diminished 
and frequent additions of profit to price avoided. By the 



A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. 171 

acquisition of an independent business capital the artisan 
class is changed from a mere wage-earning class of labour- 
ers into a capitalistic producing class; and the movable 
property now, dissociated from land-ownership, accumu- 
lates in its hands and becomes the basis of an independent 
social and political reputability which is embodied in the 
burgher class. 

The direct relationship between the handicraftsman and 
the consumer of his products makes it necessary that the 
business remain small. Whenever any one line of handi- 
craft threatens to become too large, new handicrafts split 
off from it and appropriate part of its sphere of produc- 
tion. This is the mediaeval division of labour,**^ which con- 
tinually creates new and independent trades and which led 
later to that jealous delimitation of the spheres of work 
that caused a large portion of the energy of the guild 
system to be consumed in internal bickerings. 

Handicraft is a phenomenon peculiar to the town. Peo- 
ples which, like the Russians, have developed no real town 
life, know likewise no national handicraft. And this also 
explains why, with the formation of large centralized 
States and unified commercial territories, handicraft was 
doomed to decline. In the seventeenth and eighteenth 
centuries there was developed a new industrial system, 
based no longer on the local but on the national and inter- 
national market. Our ancestors have denoted this sys- 
tem by the two names manufactories and factories, without 
distinguishing between the two terms. When viewed 
more closely these are seen to indicate two quite distinct 
industrial systems. The one hitherto characterized by the 
misleading phrase house industry we prefer to call the com- 

" For details sec my work, Die Bevolkerung von Frankfurt a. M. itn 
XIV. und XV, Jahrhundert, I, p. 228. Compare also Chapters III and 
VIII. 



172 A HISTORICAL SURyEY OF INDUSTRIAL SYSTEMS, 

mission system,^^ the other is our factory system. Both 
systems undertake the work of supplying a wide market 
with industrial products, and both require for this purpose 
a large number of labourers; they differ only in the man- 
ner in which they accomplish the work and organize the 
labourers. 

In this respect the method of the commission system 
is the simplest. In the first place, it leaves the exist- 
ing method of production quite undisturbed and confines 
itself to organizing the market. The business undertaker 
is a commercial entrepreneur who regularly employs a 
large number of labourers in their own homes, away from 
his place of business. These labourers are either former 
handicraftsmen who now produce for a single tradesman 
instead of for a number of consumers, or former wage- 
workers who now receive their raw material, not from the 
consumer, but from the merchant; or, finally, they are 
peasant families, the former products of whose domestic 
work are now produced as market wares and by the en- 
trepreneur introduced into the markets of the World. 

In some cases the entrepreneur advances ^® to the small 
producers, who at first enjoy a fairly independent position, 
the purchase price of their products; in some cases he fur- 
nishes them with the raw material, and then pays piece- 
work wage; while in others he owns even the principal 
machinery, such as the weaver's loom, the embroidering 
machine, etc. As the small producers have only the one 
customer they gradually sink into ever-greater depend- 
ence. The entrepreneur becomes their employer, and they 
are employees, even when they supply the raw material 
themselves. 

It is scarcely necessary to describe in detail the 

" Verlag. 

" Verleger comes from Verlag, i.e, supplying or advancing. 



A HISTORICAL SURVEY OF IHDU STRIA L SYSTEMS. I73 

commission system and its contingent method of work, 
house industry. We have plenty of examples in the moun- 
tain districts of Germany, for instance, the straw-plating 
and the clock and brush industries in the Black Forest, the 
wood-carving of Upper Bavaria, the toy manufacture in 
the Meiningen Oberland,the embroidery of the Voigtland, 
the lace-making of the Erzgebirge, etc. The history and 
present condition of these industries have been fairly well 
investigated in recent times. But we can no more enter 
into them than into the great variety of phases presented 
by this form of industry. 

The essential feature is ever the transformation of the 
industrial product, before it reaches the consumer, into 
capital — that is, into a means of acquisition for one or more 
intermediary merchants. Whether the entrepreneur place 
the product on the general market, or keep a town ware- 
room from which to sell it; whether he receive the wares 
from the houseworker ready for sale, or himself subject 
them to a last finishing process; whether the workman call 
himself master and keep journeymen, or whether he be a 
tiller of the soil as well — the house workman is always far 
removed from the real market of his product and from a 
knowledge of market conditions, and therein lies the chief 
cause of his hopeless weakness. 

If under the commission system capital has merely as- 
sumed control of the marketing of the products, under the 
factory system it grasps the whole process of production. 
The former system, in order to accomplish the productive 
task falling to it, draws loosely together a large number of 
homogeneous labourers, imparts to their production a defi- 
nite direction, approximately the same for each, and 
causes the product of their labour to flow, as it were, into 
a great reservoir before distributing it in all directions. 
The factory system organizes the whole process of pro- 



174 A HISTORICAL SURyEY OF INDUSTRIAL SYSTEMS, 

duction; it unites various kinds of workers, by mutual 
relations of control and subjection, into a compact and 
well-disciplined body, brings them together in a special 
business establishment, provides them with an extensive 
and complex outfit of the machinery of production, and 
thereby immensely increases their productive power. The 
factory system is as distinguishable from the commission 
system as the well-organized, uniformly equipped regular 
army from the motley volunteer militia. 

Just as in an army corps ready for battle, troops of 
varied training and accoutrement — infantry, cavalry, and 
artillery regiments, pioneers, engineers, ammunition col- 
umns and commissariat are welded into one, so under the 
factory system groups of workers of varied skill and equip- 
ment are united together and enabled to accomplish the 
most difficult tasks of production. 

The secret of the factory's strength as an institution 
for production thus lies in the effective utilization of labour. 
In order to accomplish this it takes a peculiar road, which 
at first sight appears circuitous. It divides as far as pos- 
sible all the work necessary to a process of production into 
its simplest elements, separates the difficult from the easy, 
the mechanical from the intellectual, the skilled from the 
rude. It thus arrives at a system of successive functions, 
and is enabled to employ simultaneously and successively 
human powers of the most varied kind — trained and un- 
trained men, women and children, workers with the hand 
and head, workers possessing technical, artistic and com- 
mercial skill. The restriction of each individual to a small 
section of the labouring process effects a mighty increase 
in the volume of work turned out. A hundred workmen 
in a factory accomplish in a g^ven process of production 
more than a hundred independent master craftsmen, al- 
though each of the latter understands the whole process, 



A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. I75 

while none of the former understands more than a small 
portion of it. As far as the struggle between handicraft 
and factory is fought out on the ground of technical skill, 
it is an evidence how the weak overcome the strong when 
guided by superior intellectual power. 

The machine is not the essential feature of the factory, 
although the subdivision of work just described has, by 
breaking up the sum of labour into simple movements, end- 
lessly assisted and multiplied the application of machinery. 
From early times machines for performing tasks and for 
furnishing power have been employed in industry. In 
connection with the factory, however, their application 
attained its present importance only when men succeeded 
in securing a motive power that would work unintermit- 
tently, uniformly and ubiquitously, namely, steam; and 
even here its full importance is felt only in connection 
with the peculiar industrial form of factory manufacture. 

An example will serve to illustrate what has just been 
said. In the year 1787 the canton of Zurich had 34,000 
male and female hand-spinners producing cotton yam. 
After the introduction of the English spinning-machines 
a few factories produced an equal or greater quantity of 
thread, and the number of their workers (chiefly women 
and children) fell to scarcely a third of what it had been 
before. What is the explanation? The machines? But 
was not the then-existing spinning-wheel a machine? Cer- 
tainly it was; and, moreover, a very ingenious one. Ma- 
chine was thus ousted by machine. Or better, what had 
hitherto been done by the woman hand-spinner with her 
wheel was now done by successive collaboration of a whole 
series of various kinds of workers and machines. The 
entire spinning process had been decomposed into its sim- 
plest elements, and perfectly new operations had arisen for 
which even immature powers could in part be utilized. 



176 A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. 

In the subdivision of work originate these further pe- 
culiarities of factory production— the necessity of manu- 
facture on a large scale, the requirement of a large capital, 
and the economic dependence of the workman. 

With regard to the two last points we easily perceive 
an important difference between the factory and the com- 
mission system. Its large fixed capital assures to factory 
work greater steadiness in production. Under the com- 
mission system the house-workers can at any moment be 
deprived of employment without the entrepreneur running 
any risk of losing capital; but the manufacturer must in 
like case go on producing, because he fears loss of inter- 
est and shrinkage in the value of his fixed capital, and 
because he cannot afford to lose his trained body of work- 
men. This is the reason why it is probable that the com- 
mission system will long maintain itself alongside factory 
production in those branches of industry in which the 
d\gmand is liable to sudden change, and in which the articles 
produced are of great variety. 

If, in conclusion, we were briefly to characterize these 
five industrial systems, we might say that housework is 
production for one's own needs, wage-work is custom 
work, handicraft is custom production, commission work 
is decentralized, and factory labour centralized production 
of wares. As no economic phenomenon stands iso- 
lated, each of these systems of industry is at the same time 
but a section of a great economic and social order. House- 
work is the transformation of materials in the autonomous 
household economy; wage-work belongs to the period of 
transition from independent household economy to town 
economy; the hey-day of handicraft coincides with the 
period when town economy reached its full development; 
the commission system is a connecting link between town 
economy and national economy (independent State econ- 



 

4 



A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. I77 

omy), and the factory system is the industrial system of 
fully developed national economy. 

It would lead us too far to explain in this chapter how 
each industrial system fits organically into the contem- 
porary method of production and how it is mutually deter- 
mined by a series of allied phenomena in the spheres of 
agriculture, personal services, trade and transportation. It 
can scarcely escape the observant eye that all the elements 
of the evolution here broadly sketched are contained in the 
primitive cell of society, the family; or, in economic phrase, 
in the conditions of production in the independent house- 
hold. From this primitive social unit, teeming with life and 
swallowing up all individual existence, parts have continu- 
ally detached themselves through differentiation and in- 
tegration, and become more and more independent. Wage- 
work is only a sprout from the root of the tree of independ- 
ent household economy; handicraft still needs its protec- 
tion in order to flourish; commission work makes the mar- 
keting of products a special business, while production 
sinks back almost to the first stages of development. Fac- 
tory manufacture, on the other hand, permeates with the 
entrepreneur principle the whole process of production; 
it is an independent economic system freed from all ele- 
ments of consumption, and separated as regards commod- 
ities and locality from the household life of those engaged 
in it. 

The position of the worker changes in a similar way. 
With the commencement of wage-work the industrial 
worker separates himself personally from the independent 
household economy of the landed proprietor; with the 
transition to handicraft he also becomes, through the elim- 
ination of business capital, materially free and independent. 
Through the commission system he enters into a fresh per- 
sonal subjection, he falls into dependence upon the capi- 



178 A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. 

talistic entrepreneur; under the factory system he becomes 
also materially dependent upon him. By four stages of 
evolution he passes from manorial servitude to factory 
servitude. 

There is a sort of parallelism in this evolution. The re- 
lation between the unfree houseworker and the ancient 
landowner bears a certain resemblance to the relation be- 
tween the factory hand and the modern manufacturer; and 
the wage-worker occupies much the same position with 
regard to the economy of the landed proprietor that the 
worker engaged in house industry does to the entrepreneur 
giving out commission work. In the middle of this ascend- 
ing and descending series stands handicraft as its founda- 
tion and comer-stone. From housework to handicraft we 
see the gradual emancipation of the worker from the soil 
and the formation of capital; from handicraft to the factory 
system a gradual separation of capital from work, and the 
subjection of the worker to capital. 

At the stage of housework capital has not yet emerged; 
there are only consumption goods at various stages oi 
ripeness. Everything belongs to the household — raw ma- 
terial, tools, the manufactured article, often the worker 
himself. In the case of wage-work the tools are the only 
capital in the hands of the worker; the raw and auxiliary 
materials are household stores not yet ready for consump- 
tion; the work-place belongs, under the system of migra- 
tory labour, to the domestic establishment that is to con- 
sume the finished product, or, under the housework sys- 
tem, to the worker who produces the article. In the case 
of handicraft the tools, work-place, and raw material are 
capital in the possession of the worker; the latter is master 
of the product, though he invariably sells it to the imme- 
diate consumer. In the commission system the product 
also becomes capital — not the capital of the worker, how- 



A HISTORICAL SURVEY OS INDUSTRIAL SYSTEMS. I79 

ever, but of quite a new figure on the scene, the commercial 
entrepreneur; the worker either retains all his means of 
production, or he loses possession successively of his goods, 
capital, and his implements of production. Thus all the 
elements of capital finally unite in the hand of the manu- 
facturer, and serve him as a foundation for the reorganiza- 
tion of industrial production. In his hands even the 
worker's share in the product becomes a part of the busi- 
ness capital. 

This share of the worker consists, at the stage of house- 
work, in a participation in the consumption of the finished 
products; in the case of wage-work it consists in board, 
together with a time- or piece-work wage, which even at 
this point includes compensation for wear and tear of tools; 
in handicraft it consists in the full returns from production. 
Under the commission system the commercial undertaker 
takes away a portion of the latter as profit on his business 
capital; under the factory system all the elements of pro- 
duction which can be turned into capital become crystal- 
lizing centres for further profits on capital, while for the 
worker there remains only the stipulated wage. 

We must not, however, imagine the historical evolution 
of the industrial system to have been such that each new 
industrial method absolutely superseded its predecessor. 
That would be just as far astray as, for example, to suppose 
that a new means of communication supplants those 
already existing. Railways have done away neither with 
conveyances on the highways, nor with transportation by 
means of ships, pack-animals or the human back; they 
have only confined each of these older methods of trans- 
portation to the field in which it can best develop its pe- 
culiar advantages: it is probable that not only abso- 
lutely but relatively more horses and men are employed in 



y 



1 80 A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS, 

the work of transportation in our civilized countries to-day 
than there were in the year 1830. 

The very same causes that have produced such an enor- 
mous increase in traffic are also at work in the sphere of 
industry; and in spite of the continual improvement of the 
mechanical means of production they demand an ever-in- 
creasing number of persons. From two quarters, how- 
ever, the sphere of productive industry is constantly re- 
ceiving accessions; first, from the old household economy 
and agriculture, from which even to-day parts are always 
separating themselves and becoming independent branches 
of industry; secondly, from the continual improvement*^ 
and increase in range of articles serving for the satisfac- 
tion of our wants. 

As regards the first point, there have sprung up in the 
industrial world during the last generation dozens of new 
trades for taking over such kinds of work as used formerly 
to fall to the women of the household or to the servants, 
such as vegetable and fruit preserving, fancy baking and 
preparation of meats, making and mending women's and 
children's clothes, cleaning windows, feather beds and cur- 
tains, chemical cleaning and dyeing, painting and polishing 
floors, gas and water installation, etc. Under the heading 
"Art and Market Gardening," the latest statistics of trades 
in the German Empire give thirty-five, and under the head- 
ing '* Stock-raising," thirty-one, independent occupations, 
many of which are of very recent origin. 

With regard to the second point, we will mention only 
the bicycle industry, which within a short time has not only 

•• In reply to a criticism of this expression in the Revue d'economie 
politique for November, 1892 (p. 1228, note), we will not omit making 
it more definite by saying that we do not mean by it the improvement 
of the quality of already existing species of goods, but the supplant- 
ing of existing goods by others which better and more cheaply supply 
the demand. 



A HISTORICAL SURREY OF INDUSTRIAL SYSTEMS. i8l 

necessitated the erection of a great number of factories, 
but has already given rise to special repair-shops and sep- 
arate establishments for the manufacture of rubber tires, 
cyclometers and bicycle spokes. A still more striking ex- 
ample is afforded by the application of electricity. In the 
industrial census of 1895 there are enumerated names of 
twenty-two electrical occupations that did not exist in 1882. 
The production of electrical machines, apparatus and plant 
in the German Empire gave employment in 1895 to 14.494 
persons, with 18,449 members of their families and serv- 
ants — thus furnishing a living for nearly 33,000 persons.^^ 
In metal-work, in the manufacture of machinery, chemicals, 
paper, in the building industries, the clothing and cleaning 
industries the number of recorded occupations more than 
doubled itself between 1882 and 1895, It is, at the same 
time, to be remembered not only that specialization has 
made immense strides, but that in many instances subsid- 
iary articles of production and trade which have hitherto 
been produced by the businesses using them are the ob- 
jects of separate enterprises. In these fields industry not 
only meets demand but frequently outruns it, as has at all 
times been the case. In the patent lists we find significant 
expression of this effort to improve the world of com- 
modities; and though many of the new inventionii prove 
deficient in vitality, there always remains a considerable 
number whereby life is permanently enriched. 

" In a report appearing in the newspapers of August, 1900, Dr. R. 
Burner estimates the total capital of German firms manufacturing 
electrical apparatus, in round numbers, at 800 million marks (200 
million dollars), and the stock of the so-called financial corporations, 
which* are taken up with the laying of electric car lines and works, at 
450 million marks (112 million dollars). The electric lines, electric 
works and block stations in Germany are credited, in round numbers, 
with 1,250 million marks (312 million dollars). So that the whole 
electrical plant of Germany represents a capital of about two and a 
half milliards of marks (625 million dollars). 



7 



l82 A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. 

If we were able statistically to bring together the whole 
sum of industrial products produced yearly in Germany in 
such a way that we could separate the output of factories, 
of house industry, and of handicraft, wage-work and 
housework, we should without, doubt find that the 
greater part of the factory wares embraces goods which 
were never produced under any of the other industrial sys- 
tems, and that handicraft produces to-day an absolutely 
greater quantity than ever before. The commission and 
factory systems, it is true, have completely absorbed some 
of the lesser handicrafts and robbed many others of por- 
tions of their sphere of production. But all the great guild 
handicrafts that existed at the close of the i8th century, 
with perhaps the single exception of weaving, still exist 
to-day. Handicraft is constantly being displaced by the 
more perfect industrial systems, just as in mediaeval times 
housework and wage-work were ousted by handicraft, 
only now it occurs in a less violent manner, on the field of 
free competition. This competition of all with all, sup- 
ported as it is by a perfected system of transportation and 
communication, often compels the transition from custom 
to wholesale production, even where from the technical 
standpoint the former might still have been possible. 
Many independent master workmen enter the service of 
the entrepreneur carrying on commission or factory work 
just as their predecessors a thousand years ago became 
manorial labourers. 

Handicraft has thus been relegated economically and so- 
cially to a secondary position. But even if it will no longer 
flourish in the large towns, it has in compensation spread 
all the more in the country, and here called forth, in com- 
bination with agriculture, numerous industries upon which 
the eye of the philanthropist can rest with delight. Handi- 
craft, it may be said with certainty, will no more disappear 



A HISTORICAL SURVEY OF INDUSTRIAL SYSTEMS. 183 

than wage-work and housework have disappeared. What 
it has won for society in a time of universal feudalization, 
namely, a robust class of people independent of landed 
property, whose existence is based upon personal worth 
and a small amount of movables, and who are a repository 
of popular morality and uprightness — that will and must 
remain a lasting possession, even though the existence of 
those whom these virtues will in future adorn may rest 
upon a different basis. 

In recent times there has been raised with rare persist- 
ence a cry for the uprooting of the older industry. Handi- 
craft, house industry, in general all forms of work on a 
small scale are, we are told, a drag upon the national pro- 
ductive power; they are "antiquated, superseded, rude, not 
to say socially impeditive methods of production," which 
in the best interests of those who follow them must be re- 
placed by a " rational and judicious organization and regu- 
lation of human activities on a large scale," if the actual 
national production is not to lag far behind what is tech- 
nically possible. 

This short-sighted economico-political theorizing is not 
new. There was once a time when every peasant shoe- 
maker who raised his own potatoes and cabbage was 
looked upon as a sort of enemy to the highest possible na- 
tional wealth, and when people would have liked to force 
him by police regulation to stick to his last, even though at 
the same time he ran the risk of starving. Truly, it has al- 
ways been much easier to^ censure than to understand. 

If, instead of such dogmatic pronouricements, a willing- 
ness had been shown to make an unbiassed investigation of 
the conditions governing those older and supposedly an- 
tiquated systems of production, the conviction would soon 
have arisen that in the majority of cases where they still 
persist they are economically anfd socially justifiable; and 



l84 A, HlSTORJC/iL SURVEY OF INDUSTRIAL SYSTEMS. 

the means for the removal of the existing evils would be 
sought in the soil in which these industrial forms are rooted 
instead of such drastic remedies being applied to them. 
In this way we should undoubtedly preserve the good of 
e^ch of these individual systems and be striving only to re- 
move their disadvantages. 

For, after all, the comforting result of every serious con- 
sideration of history is, that no single element of culture 
which has once entered into the life of men is lost; that 
even after the hour of its predominance has expired, it 
continues in some more modest position to cooperate in 
the realization of the great end in which we all believe, the 
helping of mankind towards more and more perfect forms 
of existence. 



/ 



CHAPTER V. 

THE DECLINE OF THE HANDICRAFTS. 

There are in Germany two handicraft problems. One 
is a problem belonging to the newspapers and the legisla- 
tures, which since 1848 has repeatedly occupied the liveliest 
public attention. For it the question is the extent to which 
the particular interests of hand-workers as a class should be 
given legislative expression. What the answer shall be 
depends upon the relative strength of political parties. 

The other problem relates to the vitality of hand-work 
as a form of industrial activity. It is the query of Ham- 
let's soliloquy: "To be or not to be!" The answer de- 
pends upon actual conditions. Stated more explicitly, the 
question would read: How far has hand-work up to the 
present shown itself capable of holding its own? What de- 
partment of industry does it still dominate? 

So long as public policy weighs, not merely wishes and 
votes, but also the facts of the case, it will not venture to 
decide the first of these questions until the second has been 
answered. Until lately the necessary established data 
have been lacking. Recently, however, the German Social 
Science Club has conducted a most comprehensive investi- 
gation into those branches of industry belonging to the old 
class of handicrafts.* It is therefore opportune to give a 

^Investigations as to the Condition of Handwork in Germany, with 
special Reference to its Ability to compete with industrial Undertakings on 
a large Scale, in Schriften des Vereins fur Sozialpolitik, Vols. 62-70 

185 



1 86 THE DECUNE OF THE HANDICRAFTS. 

general survey of the findings. In this it is not our inten- 
tion to enter upon a discussion of the present position and 
future prospects of particular industrial branches,* but 
rather to present the common characteristics of the de- 
velopment that has taken place during the past hundred 
years or more. This will make it possible to appreciate in 
their full strength and manifold modes of operation the 
forces in modern national economy that act as solvents and 
as creative agents. 

A century ago handicraft still held undisputed sway over 
all its mediaeval inheritance and over its sixteenth and 
seventeenth century conquests as well. There existed be- 
sides, it is true, a few manufactories and factories. But they 
had developed apart from hand-work; what they produced 
had never been handicraft work. Rivalry between these 
new forms of industry and the guild hand-work there had 
never been. Nor had the guilds as such been interfered 
with by the State; they had only been made amenable 
to its laws, and thusJ^een in part stripped of their local 
municipal character. Their scope indeed had been ex- 
tended, in that those handicrafts were subjected to the 
guild constitution, which, because of the limited number of 
their representatives, had not as yet been able to form 
local guilds in the various towns. Through the territorial 
guilds, which had been constituted for these "minor handi- 
crafts," and through the " general guild articles " com- 
pactly summarizing uniform trade regulations for all local 
guilds, the requirements of modern national economy had 

(Leipzig, 1894-1897). A further volume (Vol. 71) relates to Austria. 
Supplementing this is the inquiry into the conditions of handwork, 
undertaken in the summer of 1895, and edited by the Imperial Statis- 
tics Bureau (3 numbers, Berlin, 1895). 

' The results of the investigations of H. Grandke along this line are 
presented in Schmoller's Jahrbuch fur Gesetzgeb, Verwalt. u. Volksw., 
Vol. XXI (1897), pp. 1031 ff. 



THE DECLINE OF THE HANDICRAFTS. 187 

been at least formally asserted. In practice, however, the 
local and craft prerogatives of sale, the town monopoly 
and extra-mural rights of jurisdiction ^ remained in force. 
Competition between the members of the same handicraft 
from different towns and of the different crafts of the same 
towns was entirely lacking; settlement in rural parts was 
for most crafts forbidden, and to gain independence was 
made a matter of extreme difficulty for all journeymen 
who were not masters* sons or sons-in-law. 

But what was the condition of the master craftsmen in 
exclusive possession of these privileges? 

Most of those who discuss handicraft to-day depict the 
masters " of the golden era of handicraft " as well-to-do 
people carrying on business " with considerable capital for 
those times," owning " their own dwellings and extensive 
workshops," working along with master journeymen and 
apprentices, personally capable, honourable, respected. All 
the painters dip their brush in glowing colours, so neces- 
sary for the portrayal of a condition of prosperity. 

Whence have they this picture? We have vainly sought 
it in the eighteenth or seventeenth century. Moreover, 
our classical poets could not have had it before their eyes; 
for their " gossiping tailors and glovers " are petty, insig- 
nificant apparitions. In the multitude of small towns the 
masters were able to maintain themselves only through 
their bit of farming and the lucrative brewing privilege, 
and in the larger towns through the little counter k^pt by 
many of them in connection with their workshop. Even 
for a town of such commercial prominence as Leipzig, the 
mass of administrative records from the last two centuries 
do not allow the impression that the craftsmen of that 
place were on the average well off; and the extensive liter- 

 [Comp. on the latter Roscher. System der Volksw., 5th ed., Vol. 3, 
i 128.— Ed.] 



i88 THE DECLINE OF THE HANDICRAFTS. 

ature on guilds that has come down from the close of the 
last and the beginning of the present century, the " Pa- 
triotische Phantasien " of Justus Moser, points in many in- 
stances to very narrow circumstances. 

The barriers erected against admission to mastership, 
though extensive, had not been successfully defended. 
Among the bakers and butchers, whom it is customary to 
cite as types of prosperity, baking and killing in rotation 
was the almost universal rule; that is, there were so many 
masters that each baker could not bake afresh each day, 
nor each butcher kill a head of cattle per week. As late 
as 1817 a writer cites as a normal case from Bavaria that 
in a town with ten master bakers three bakings of bread 
were consumed daily, so that every week the turn fell to 
each twice. The butchers could slaughter regularly only 
the smaller kinds of stock. In North German towns mat- 
ters appear to have been in a favourable condition if one 
beef were sold every week for every five or six masters. 

Almost all crafts with a guild organization had a clause 
in their statutes fixing the maximum number of journey- 
men and apprentices which a master might keep. As a rule 
he was limited to two. In the i8th century this number was 
rarely exceeded. Under normal conditions, however, the 
great majority of trades could not attain to this number. 
Assuming that all who learned the handicraft acquired 
master's standing, that a master lived on the average thirty 
years after he attained that rank, and that ordinarily a man 
became independent between the twenty-eighth and thir- 
tieth year of his life, there could not have been at any time 
more than half as many journeymen and apprentices as 
masters. 

The actual proportion at the end of the century was 
often much smaller still. In the year 1784 there were in 
the duchy of Magdeburg 27,050 independent masters and 



THE DECLINE OF THE HANDICRAFTS. 189 

only 4,285 assistants and apprentices. About the same 
time, in the principality of Wiirzburg (in Bavaria), 13,762 
masters with 2,176 assistants and apprentices were re- 
turned.* In both territories there were for every hundred 
masters but 15.8 journeymen and apprentices. Thus, if 
we assume that the assistants were equally distributed 
among the masters, one journeyman or apprentice hardly 
fell to each sixth master. In more than five-sixths of the 
instances the master carried on his work single-handed. 
In 1780 the town of Bochum (in Westphalia) counted for 
every five master masons one; in the other crafts they were 
for every twenty-six master shoemakers three, for every 
twenty-one master bakers, every eight carpenters and 
every five master masons one; in the other crafts they 
were altogether lacking. 

In some parts of Prussia, especially in Berlin, the con- 
ditions were indeed somewhat more favourable. But in 
general the idea must be abandoned that our modern in- 
dustrial development began with the handicrafts in a con- 
dition of general prosperity. The best that past times 
could offer the craftsmen was a modest competence, se- 
curity against lack of work and against over-severe com- 
petition from their fellows. They deal directly with their 
customers, in quiet times work up a stock and take it to 
market, and stand firmly together in the guild if it is a 
question of voting down a new application for mastership, 
taking action against an itinerant workman or resisting 
an encroachment on the part of a neighbouring craft. To- 
wards one another, however, they are possessed with the 
pettiest bread-and-butter jealousy, and give a great deal of 
trouble to the courts and administrative officials. Such 
was the early handicraft. 

^According to Schmoller, Zur Gesch. d, deutsch Kleingewerbe im 
jg. Jhdt, pp. 21, 22. 



I90 THE DECLINE OF THE H/iNDlCR/IFTS. 

Down to the fourth decade of last century there was 
no really great change. After the time of Napoleon, the 
old industrial policy was repeatedly moderated, but in most 
parts of Germany it was not abolished until the sixties. It 
gave place to industrial freedom. Anyone might now carry 
on any business anywhere and on any desired scale. The 
local prohibitive powers fell to the ground. Each trades- 
man could dispose of his products where he would, and in 
his own locality had to tolerate all external competition. 
The barriers between the different branches of industry 
dropped away, and everyone could manufacture what was 
to his advantage. 

All this took place with the full assent of the craftsmen 
themselves. The conviction that the old industrial polity 
had become untenable was shared in at least the more ad- 
vanced parts of Germany by all. If ever an old insti- 
tution was abolished with the approbation of the whole 
nation, it was the guild system. The sole sporadic mis- 
giving was that apprenticeship might fall into decay, and 
that many would establish themselves as independent 
craftsmen who had not regularly learned their trade. This 
apprehension has proved groundless. According to the 
results of the inquiry made into handicraft in various 
districts of Germany, ninety-seven per cent, of those con- 
sidered as still carrying on an independent handicraft had 
enjoyed a preparatory training as hand-workers. The 
small remainder consisted mainly of those who had re- 
ceived their technical training in apprentice shops and 
technical schools, asylums for the blind, institutions for the 
deaf and dumb, in prison and in barracks. 

The influence of the new conditions on the number, local 
distribution, and extent of undertakings assumed a differ- 
ent fornl. At first it was feared that the establishment of 
numerous petty master workmen without capital would 



THE DECLINE OF THE HANDICRAFTS. 191 

lead to a mass of half-developed business undertakings; 
but this has in nowise proved the case. On the contrary, 
after a brief transitional period, the undertakings in the 
towns have on the average during the last generation 
numerically diminished, while in financial strength and in 
the number of assistants, so far as their existence has not 
been in general jeopardized by causes lying outside the 
province of industrial legislation, the respective branches 
of trade have increased. At the same time hand-work has 
made striking advances in the country, and to-day is ap- 
proximately as strong there as in the towns. 

This equalization between town and country, however, 
had been foreseen and aimed at at the time by the advo- 
cates of industrial freedom. If there was the further 
expectation that through industrial freedom the way of 
the artisan would be opened to technical progress and to 
economic advancement, this also has not lacked fulfilment. 
Thousands of urban master workmen have in the last two 
generations become large manufacturers, or at least capi- 
talistic entrepreneurs, and have participated fully in the 
technical advances of the period. Freedom of industry 
and enterprise has made it possible for them to broaden 
their sphere of production and sale, and to utilize fully their 
personal ability. This fact men to-day are only too prone 
to overlook. 

To be sure, the number of those who have not risen in 
the world, but remained stationary, and of those reduced 
to the level of master jobbers and homeworkers, or forced 
to become factory employees, is much larger still. Whole 
branches- of industry formerly carried on as hand-work are 
almost ruined, or at least are lost to handicraft as a par- 
ticular industrial form. Others are still struggling for their 
existence. A great weathering and transforming process 
has here come into operation; in its train handicraft is 



192 THE DECLINE OF THE HANDICRAFTS. 

yielding place to other forms of business, such as the fac- 
tory and commission systems, or the hybrid forms that 
every period of transition begets. 

The public at large is content to include all involved in 
these processes under the simple headings: displacement 
of hand-work by machinery, annihilation of handicraft by 
the factory! The smaller cost of production by machinery 
is looked upon as the sole cause. 

The reduction of these expressions to their true value, 
and the demonstration that a large part of the changes 
which have taken place has its cause not in the advances of 
manufacturing technique but in the direction taken by 
economic consumption^ and that so far as this is the case 
handicraft disappears even without machine-work coming 
into competition with it — this will remain one of the 
greatest services rendered by recent investigators of hand- 
work. It will be necessary first to present a summary view 
of these changes in consumption, since they, so to speak, 
condition the whole development. 

In the first place, a local concentration of demand has taken 
place. The aggregations of human beings that have been 
formed in great cities in the course of the last half cen- 
tury, furthermore the standing armies, the large state and 
municipal institutions, prisons, hospitals, technical schools, 
etc., the extensive establishments for transportation, fac- 
tories, and large undertakings in the departments of trade, 
banking and insurance, all form centres of wholesale de- 
mand for industrial products. To these are to be added 
the great departmental warehouses, export businesses and 
cooperative societies, focussing the demand of large sec- 
tions of the population at a few points. This demand they 
are no longer able to satisfy as customers of individual 
craftsmen. 

There comes then as a second consideration the many 



THE DECLINE OF THE HANDICIOIFTS. 193 

instances in which modern civilization has propounded 
such colossal tasks for industry that they cannot be accom- 
plished at all with the implements and methods of handi- 
craft, although each of them generally requires consid- 
erable hand-work. The manufacture of a locomotive, of a 
steam crane, of a rapid press, the building of a river bridge 
or of a warship, the equipment of a street railway with 
rails and rolling stock cannot be carried out with mere 
hand apparatus and manual labour. They require im- 
mensely powerful mechanical appliances, highly trained 
engineers and craftsmen of exceedingly varied qualifica- 
tions. 

Even where technically such tasks might still be accom- 
plished with the implements of hand-work, the entrusting 
of them to master craftsmen is economically impossible be- 
cause of the consequent heavy loss of interest. In the 
Middle Ages the building of a cathedral might occupy two 
or three generations, indeed, several centuries. Imagine 
one to-day wishing to take as much time for the erection 
of a railway station! When in 1896 the contract for the 
main building of the Saxon-Thuringian Industrial Exhibi- 
tion in Leipzig was to be let, it was first offered to the 
master carpenters of the city — contractors who carry 
on work with considerable capital and are accustomed to 
large undertakings. But all hesitated because of the short- 
ness of the term for building and the extent of the risk. 
Negotiations were thereupon entered into with a large firm 
of builders in Frankfurt-on-Main. In a few hours the con- 
tract was closed. The same evening the telegraph was 
working in all directions. A week later the steam rams 
were busy on the building site, and whole trains were ar- 
riving from Galicia with the necessary timber. 

In fact, one can say that to-day there are industrial tasks 
of such magnitude that they can be performed by only a 



194 THE DECLINE OF THE HANDICRAFTS. 

few, perhaps indeed by but one or two firms in Europe. 
Hence the development, beside the earlier type of factory, 
which finds its strength in the wholesale production of 
similar articles, of a new type whose raison d'etre lies in 
the magnitude of the task of production. This more recent 
kind of large industrial undertaking we might designate 
by the already current expression, manufacturing estab- 
lishment.^ At the head stands a staff of technically trained 
men, with extensive mechanical appliances at command, 
and with the necessary hand-work in most effective com- 
bination. 

But the demand for industrial labour has been not 
merely locally concentrated and condensed to meet the 
extensive requirements of production; it has also become 
more uniform, and therefore more massive. A tendency 
towards uniformity runs through our age, eliminating the 
differenc*es of habits and customs in the various strata of 
society. Characteristic peasant costumes have disappeared 
down to unimportant survivals; the furnishing of the dwell- 
ing, of the kitchen, has become, it is true, more extensive, 
but likewise more uniform. Even in the smallest home 
one finds the petroleum lamp, the coffee-mill, some enam- 
elled cooking utensils, a pair of framed photographs. 
To make the desired ware accessible to the poorer classes, 
it must be easily and cheaply produced. If an article is 
lifted on the crest of a wave of fashion, the demand for it 
in a cheap form advances even up to the better situated 
grades of society, and thus the outlay for the folly of fash- 
ion is made endurable. In this way there arises a large 
demand for cheap goods for whose manufacture the earlier 
type of factory is naturally adapted. Hand-work is for such 
too expensive; where it remains technically possible it must 

* FabrikationsanstaU. 



THE DECLINE OF THE HANDICRAFTS. 19S 

be extremely specialized, and then it necessarily loses the 
ground of custom work from beneath its feet. 

There is finally another consideration to be alluded to, 
which belongs to the sphere of domestic economy. The home 
is being relieved more and more of the vestigial elements 
of production, and is restricting itself to the regulation of 
consumption. If our grandparents required a sofa, they 
first had the joiner make the frame, then purchased the 
leather, the horsehair and the feathers, and had the up- 
holsterer finish the work in the house. The procedure was 
similar for almost every more important piece of work. 
To-day specialized work demanding the whole strength of 
each individual, frequently to exhaustion, no longer per- 
mits such a participation in production. We will and must 
purchase what we need ready-made. We desire to be 
quickly supplied, and preferably renounce idiosyncrasies 
of personal taste, rather than undertake the risk of order- 
ing from different producers. Industry has to adapt itself 
accordingly. 

The same evolutionary process also asserts itself in de- 
partments where the individual craftsman had been accus- 
tomed from time immemorial to supply finished wares. 
Here again the modern city consumer will no longer trade 
directly with him by ordering the single piece that he re- 
quires. He is averse to waiting; he knows that often the 
work does not turn out as desired, and prefers to choose 
and compare, before he buys. 

Thus the craftsman can no longer remain a custom 
worker even in those departments in which technically he 
is fully able to cope with the demands of production. He 
no longer works on individual orders, but exclusively for 
stock — which formerly he did only in case of necessity. 
To reach the consumer he needs the intervention of 
the store. By the discontinuance of personal contact be- 



19^ THE DECLINE OF THE HANDICRAFTS. 

tween producer and consumer, hand-work as a phase of in- 
dustry disappears. It becomes a capitalistic undertaking, 
and demands management in accord with mercantile prin- 
ciples. All now depends upon the question whether busi- 
ness on a large or a small scale offers the greater advan- 
tages. In the first case the department of work formerly 
represented by handicraft falls to the factory, in the latter 
to domestic industry. 

For even where modern demand has not yet appeared 
as wholesale concentrated demand, or*become condensed 
to meet the necessities of production on a grand scale, it 
is universally well adapted, by virtue of its great uniformity 
and its emancipation from household labour, to localization 
at a few points. The perfected commercial machinery of 
modern times, the low tariffs for post and telegraph, the 
rapidity and regularity of freight and news transportation, 
the innumerable means of advertising and of making an- 
nouncements afford here their mighty assistance. Indus- 
trial freedom thus found a well-prepared soil when it 
sprang into life. It but created the legal forms that voice 
the character of modern economic demand. All those 
circles of consumers of the craftsmen so long kept arti- 
ficially asunder could now be united through the interven- 
tion of commerce into a large manufactory and commis- 
sion clientele, not necessarily Hmited to national boundaries. 

Concentrated demand does not permit of satisfaction by 
scattered production. Along with the process of concen- 
tration of demand must go a process of concentration in the 
department of industrial production. It is to this that handi- 
craft on every side succumbs. 

But this process is very complicated, and it is not alto- 
gether easy to separate from one another the individual 
processes of which it is composed. We will nevertheless 
essay the task, choosing the fate of hand-work as the deter- 



THE DECUNE OF THE HANDICRAFTS, 197 

mining factor in the divisions made by us. We thus arrive 
at the five following cases: 

1. Supplanting of hand-work by similar factory produc- 
tion. 

2. Curtailment of its department of production by fac- 
tory or commission. 

3. Incorporation of hand-work with the large undertak- 
ing. 

4. Impoverishment of hand-work by shifting of demand. 

5. Reduction of hand-work by way of the warehouse to 
home and sweat-work. ^ 

Several of these processes often go on simultaneously. 
In our consideration of the subject, however, we will keep 
them as far as possible apart. 

I. The case in which capitalistic production on a large 
scale attacks handicraft along its whole front, in order to 
expel it completely from its sphere of production is compara- 
tively rare. From earlier times we may mention weaving, 
clock and gun making, and also the smaller industries of 
the pin-makers, button-makers, tool-smiths, card-makers, 
hosiers; from recent times hatmaking, shoemaking, dyeing, 
soap manufacture, rope-making, nail and cutlery smithing, 
comb-making: to a certain extent beer-brewing and 
coopering also belong to the list. 

The process of displacement assumes now a quicker now 
a less rapid character, according as the handicraft in ques- 
tion formerly carried on manufacture for stock along with 
market and shop sale, or restricted itself to custom work. 
Thus the making of shoes for market sale paved the way 
for the manufacture of shoes by machinery, because it had 
long accustomed certain classes of the people to the pur- 
chase of ready-made footwear. 

For handicraft the result of such a development varies 
according as the factory product, after being worn out, 



Z' 



198 THE DECLINE OF THE HANDICRAFTS. 

permits of repair or not. In the latter case handicraft dis- 
appears altogether; in the former it evolves into a repair 
trade, with or without a sale shop. The carrying on by a 
hand-worker of a shop trade with factory goods in his own 
line is not exactly an unfavourable metamorphosis; but 
only craftsmen with considerable capital can manage it. 
On the other hand, pure repair trade very easily loses 
the ground of hand-work beneath its feet, if the factory 
product passes completely into the control of retail mer- 
chants. For then the majority of consumers prefer to have 
repairs made in the shop in which they have purchased the 
new ware. The proprietor of the shop keeps a journey- 
man or sends out the mending to a petty master workman. 
This greatly diminishes their return, and makes them com- 
pletely dependent. Moreover, repairing can also be car- 
ried on on a large scale, as with the so-called rag-dyeing, 
which works with considerable capital and independent 
collecting points. Finally, the repairing can become quite 
superfluous through very cheap production of new wares, 
as, for example, with clocks and shoes; repair would cost 
more than a new article. 

2. Much more frequently does the second group of evo- 
lutionary processes make its appearance. Here it is not a 
question of the complete loss of the new manufacture, but 
merely of the curtailment of the department of production fall- 
ing to handicraft through factory or commission business. 
The causes of this process may be very diverse. While 
recognising the impossibility of being exhaustive, we will 
distinguish four of them: 

(a) Various handicrafts are fused ifito a single manufac- 
turing establishment: for example, joiners, wood-carvers, 
turners, upholsterers, painters, lacquerers into a furniture 
factory; wheelwrights, smiths, saddlers, glaziers into a car- 
riage manufactory'; basket-makers, joiners, wheelwrights, 



THE DECUNE OF THE HANDICRAFTS, 199 

saddlers, smiths, locksmiths, lacquerers into a baby-car- 
riage factory. We may mention further all kinds of ma- 
chine-shops, locomotive and car-works, piano factories, 
trunk factories, billiard-table factories, and also the estab- 
lishments for the production of whole factory plants — dis- 
tillery, brewery, sugar-refinery, etc. As a rule the part of 
production withdrawn from the individual handicraft 
through such an incorporation forms but a small fragment 
of its previous sphere of work and of its market. If, how- 
ever, such blood-lettings are frequent, as among the 
turners, saddlers and locksmiths, there finally remains 
very little, and the handicraft may die of exhaustion. 

{b) Various remunerative articles adapted to wholesale pro- 
duction by factory or house industry are withdrawn from 
hand-work. Thus bookbinding has had to resign almost its 
whole extensive department of production to more than 
forty kinds of special trades; there remains but the indi- 
vidual binding for private customers. Basket-making has 
surrendered the fine wares to homework, baby-carriages 
and the like to factories, and only the coarse willow wicker- 
work remains to handicraft. The locksmith has even lost 
the article, the lock, from which he has his name; the 
brush-maker the manufacture of paint, tooth, and nail 
brushes; the cabinet-maker has been compelled to re- 
nounce the intermediate wares (Berlin furniture), and 
ordinary pine furniture has become a stock-in-trade of 
the store; confectionery is threatened, in the cities at 
least, with being despoiled by the factories of the manu- 
facture of bread; the tinsmith no longer makes his vessels; 
in short there are likely but few handicrafts that have not 
similar losses to record. 

(c) The factory takes over the primary stages of production. 
It was precisely the first rough working of the material 
which demands the greatest expenditure of strength, it 



200 THE DECLINE OF THE HANDICRAFTS. 

was exactly this primary handling that suggested the ap- 
plication of machinery, while the finer and individual 
shaping of the product in the later stages of the process 
of production tempted the entrepreneur but slightly. In 
almost all metal and wood industries the raw material is 
now used only in the form of half-manufactured wares. 
The furriers work up skins already prepared, the smith 
purchases the finished horseshoe, the glazier ready-made 
window-frames, the brush-maker cut and bored wooden 
parts and prepared bristles, the contracting carpenter in- 
laid flooring cut as desired and doors all ready to hang. 

At first such a loss is generally felt by the handicraft con- 
cerned as an alleviation rather than an injury. The process 
of production is shortened; the individual master can pro- 
duce a greater quantity of finished articles than formerly; 
and if he reckons on each piece the same profit as for- 
merly,* his income can easily advance provided he retains 
sufficient work. A locksmith, who procures all door- 
mountings ready-made from the hardware shop, can 
readily finish several buildings in one summer, while pre- 
viously, when he had first to make these wares, he perhaps 
completed only one. But still, in most cases, through such 
a cutting into the roots of hand-work, not a few of the mas- 
ter craftsmen become superfluous. At the same time, how- 
ever, the amount of business capital required increases, 
since the craftsman has now to make disbursements not 
merely for the raw material, but also for the costs of pro- 
duction of the partly manufactured product, and further- 
more, has to furnish the manufacturer's and trader's profits 
as well. 

This is all the more vital, since just in the first-hand pur- 
chase of the raw material and in its proper selection the 
greatest profit is often made. For this reason trading 
houses have not infrequently taken over the preparatory- 



THE DECUNE OF THE HANDICRAFTS. 201 

Stages in production even where a partial manufacture 
with machinery is not to be thought of. There is abso- 
lutely no doubt that the hand-worker in wood was in a 
better position when he could purchase his wood in the 
form of logs in the forest than now, when he procures it in 
the form of boards, laths, and veneers from lumber-dealers; 
and that the brush-maker worked to greater advantage 
when he bought the rough bristles from the butcher than 
now, when he must buy them arranged by the dealer in 
innumerable classes. 

Of course this trade in partly finished goods is very con- 
venient for the craftsman; he can obtain from the dealer 
even the smallest quantities. But it is exactly this that has 
contributed not a little to the decline of the handicrafts, 
since the journeyman can now go into business almost 
without capital. Thus, for instance, in the shoe trade the 
manufacture of vamps at first greatly promoted business 
on a small scale, not because it shortened the manufactur- 
ing process for the shoemaker, but because it placed him 
in a position to purchase a single pair of uppers at the 
shoefinder's where formerly he had to procure from the 
tanner at least a whole skin. 

This cooperation of mechanical preparatory work and 
handicraft assumes a particularly interesting form where 
the whole productive part of the labour process drops 
away from hand-work. The craftsman can^then continue 
to maintain himself only if the product needs to be set in 
pla<:e or fitted. But he sinks back once more almost to the 
state of the wage-worker. Thus the locksmith and the 
joiner (the latter for ready-made doors and inlaid flooring) 
are now but " fitters "; and the role of the horseshoer nail- 
ing on ready-made horseshoes is not very different. 

On the other hand, the shortening of the process of 
manufacture makes the business more capitalistic and the 



202 THE DECLINE OF THE HANDICRAFTS. 

turnover more rapid. The vital element of handicraft, 
however, is not the profit on capital, but the labour earn- 
ings, and these under all circumstances are being curtailed. 

{d) The appearance of new raw materials and methods of 
production better adapted for manufacture on a large 
scale than those previously employed in hand-work, handi- 
caps the latter for a part of this sphere of production. 
We may cite among other instances the appearance of 
the curved (Vienna) furniture, the manufacture of wire 
nails and its influence on nail-smithing, the wire-rope 
manufacture in opposition to the hempen rope, the in- 
vasion by gutta percha of the consumption sphere of 
leather and linen. The enamelled cooking utensil has en- 
croached simultaneously upon the manufacture of pottery, 
tinsmithing and the business of the coppersmith; and the 
invention of linen for bookbinding in place of leather and 
parchment has smoothed the way for wholesale book- 
binding by machinery. 

Thus at the most diverse points handicraft is being as- 
sailed by the modern, more progressive, forms of manu- 
facture. The attacks, generally delivered in a manner to 
disarm opposition, are not infrequently made under the 
fair mask of the stronger friend taking a load from its 
shoulders, until finally nothing remains to tempt the capi- 
talistic appetite of the entrepreneur. 

3. We come now to those cases in which handicraft 
loses its independence through being appended to a large 
business. Every more extensive undertaking, be it manu- 
facturing, trading, or a general commercial establishment, 
requires for its own business various kinds of hand-work. 
As long as such tasks are few in number, they are given 
out to master craftsmen. But if they grow more numerous 
and regular, it becomes advantageous to organize a sub- 
department for them within the walls of the establishment. 



THE DECLINE OF THE HANDICRAFTS. 203 

Today every large brewery or wine-house has its own 
cooperage; the street-railway companies maintain work- 
shops for smiths, saddlers, wheelwrights and machinists; 
canning factories have their own tinshops; a shipyard 
keeps cabinet-makers and upholsterers for the internal fur- 
nishing of its passenger steamers; almost every large 
manufactory has a machine and repair shop. The master 
who enters such a large establishment as foreman of the 
special workshop ceases, of course, to be free from the con- 
trol of others, but enjoys, on the other hand, a position 
that is to a certain extent independent, and, above all, se- 
cure. 

By the free craftsmen, however, the loss of such strong 
purchasers is most bitterly felt. Indeed, the system de- 
scribed can lead to the starving out of whole crafts — a fate 
that has overtaken, for example, turning, which is being 
appended to all trades using its products in the par- 
tially manufactured state. But this process is too much in 
the interests of a good economy to make it possible to 
check it. 

The workmen for such subdepartments of a large in- 
dustrial establishment, be it further remarked, receive as 
a rule a training in their handicraft as long as it continues 
to have an independent existence. An abnormally large 
number of apprentices can thus be employed by it, while 
the journeymen have a much more extended labour market 
than the handicraft alone could offer. This is the explana- 
tion, for instance, of the occasional discovery in the lock- 
smith's trade of ten times as many apprentices as journey- 
men. 

4. Handicraft is impoverished through shifting of demand ^ 
and entirely ruined through cessation of demand. Such 
shiftings have occurred at all epochs — we may recall the 
use of parchment and periwigs — but perhaps never more 



204 THE DECUNE OF THE HANDICRAFTS. 

frequently than in our own rapidly moving times. We will 
give only a few instances. 

The cooper prepared for the household of our grand- 
parents divers vessels now sought for in vain, at least in. 
a city home: meat-barrels, tubs for sauerkraut and beans, 
washtubs, water-buckets, rain-barrels, even bathtubs and 
washing vessels. We no longer keep supplies of meat and 
preserved vegetables; water is furnished us by the water- 
works system; and the place of the small wooden vessels 
has been taken by those of tin, china, or crockery. A sec- 
ond example is offered by the turner, who formerly had to 
supply almost every household with a spinning-wheel or 
two, spools and reels. To-day the spinning-wheel has sunk 
to the position of an " old German " show-piece. Both in- 
dustries have, of course, found fresh purchasers for those 
they have lost, especially coopering, through the increase 
of barrel-packing. But the new customers are factories 
that at the earliest opportunity incorporate the cooperage 
as a subsidiary department. The industry of the pewterer 
presents a third example. The pewter plates and dishes 
that were to be found in almost every house throughout 
town and country have passed out of fashion. In their 
place have come porcelain and stoneware, and the pewter- 
er's trade has thus to all intents lost the very foundation 
of its existence. Finally, we may recall the shiftings in 
demand which the great revolutions in the sphere of travel 
have brought about, and which have fallen with especial 
severity on the saddler, trunk-maker and furrier. 

5. In a last group of instances handicraft comes into com- 
plete dependence on trade; the master becomes a home- 
worker, since his products can now reach the consumers 
only through the store. The cause of this phenomenon 
is of a double nature: on the one hand, the high rents of 
city business sites, which force the master to live and pen 



THE DECLINE OF THE HANDICRAFTS. 205 

up his workshop in a garret or a rear house where he is 
with difficulty found, and where at no time is he sought 
out by his better customers; on the other, the inclina- 
tion of the public to buy only where a larger selection is 
to be had, and where the merchant is ** accommodating," 
that is, sends goods for inspection, takes back if they do 
not suit, articles like brushes, combs, fine basket-maker's 
wares and leather goods, small wooden and metal articles 
which in larger towns are now scarcely ever purchased 
from the producer or outside the fancy-goods and hard- 
ware stores. Indeed, we even give our orders to the stores 
if we wish to have a special article made. Who to-day or- 
ders his visiting cards from the printer, or a smoker's table 
from the cabinet-maker? Anyone who has the oppor- 
tunity of seeing, along the streets that he must traverse 
perhaps several times a day, so complete a display of every- 
thing necessary for his wants that he can in a few minutes 
procure any desired article, will seldom care out of love 
for a declining handicraft to betake himself to a distant 
suburb and there, after a long inquiry and search, climb 
three or four gloomy staircases before he can deliver his 
order, in the execution of which the appointed time will 
perhaps even then be disregarded. And shall, for instance, 
anyone who finds in a furniture stock everything that is in 
any way necessary to the furnishing of a room, shall a 
young housewife who in a few hours can gather together 
in a housefumishing establishment a complete kitchen 
outfit, shall these preferably seek out a half dozen hand- 
workers from whom they can obtain what they want only 
after weeks of waiting? 

Such may be regarded as the chief features of the process 
of transformation that is taking place to-day in hand- 
icraft. We may, in conclusion, state it as a matured con- 



2o6 THE DECLINE OF THE HANDICRAFTS. 

viction compelled by the results of the investigations, that 
in all cases where it supplies finished goods which are not very 
perishable J and which can be manufactured in definite styles for 
average requirements, hand-work is endangered in the highest 
degree. This applies even where a technical superiority on the 
part of the large undertaking does not exist. These are, in 
short, cases in which the product is suited to immediate 
consumption without further assistance from the producer. 

In all these instances trade in its various branches, down 
to that of hawking, will more and more form the uni- 
versal clearing-house for industrial wares. Handicraft must 
specialise as far as possible; and it can save itself from the 
fate of dependence upon the store only by becoming a cap- 
italized industry on a small scale. The union of a sale shop 
with the workshop is then indispensable. 

In the contrary instances, where the product of handi- 
craft moist be placed in position or separately fitted, the crafts- 
man at least does not lose touch with the consumers. But 
even in such cases he can maintain himself in the large 
towns only if the demand is strongly centralized (as with 
locksmiths and generally all craftsmen connected with 
building in the widest sense), or again if he keeps a shop 
(as with tinsmithing, saddlery, or ordered tailoring), which 
serves as a collecting bureau for orders. In both cases a 
business without some capital lacks sufficient vitality to 
exist. 

With this conclusion correspond the results of the " In- 
vestigations into the Conditions of the Handicrafts." 
Everywhere in the towns the relative number of masters 
has greatly diminished, the number of their assistants in- 
creased; that is, the businesses have grown. In a still 
higher degree must their capital have advanced. Mani- 
festly it is the upper stratum of city handicraftsmen which 
has here maintained itself by adopting business methods 



THE DECLINE OF THE HANDICRAFTS. 207 

suited to the requirements of the present, and which 
probably has prospects of holding its own for some time 
to come. Where an equal variety is offered, the public 
will always prefer the shop of the master craftsman to that 
of the pure tradesman, if for no other reason, because of 
the convenience for repairs and the greater technical 
knowledge of the master. The latter, moreover, through 
the custom coming to the workshop, remains protected 
from that officious idleness to which the city shopkeeper 
so readily falls a victim. 

In the country conditions have a fairly different aspect. 
Those causes of repression of hand-work that result from 
the altered form of demand and the conditions of life in 
the towns prevail here only in a lesser degree. Rural de- 
mand is not yet so very concentrated ; it is to a large extent 
of an individual nature; everyone knows the hand- worker 
and his household personally. Connections with neigh- 
bours, school comrades or family relatives likewise play 
a part in holding trade. Here real handicraft soil is still 
to be found. The craftsman cultivates in many cases a 
bit of land; at the harvest he will assist his neighbour in 
mowing and the like; he possesses a cottage of his own; 
in short, for his sustenance he is not exclusively dependent 
upon his trade. In his business wage-work and the system 
of credit balances ® still largely prevail. 

Most of the crafts that have any real footing in the coun- 
try are in our opinion secure as far as the future can be 
forecasted. Of course, they cannot completely escape the 
revolutions in urban industry. In the country the tin- 
smith, as a rule, no longer makes the tinware he sells, 

' [For a general discussion of credit balances (Gegenrechnung) as a 
feature of public financing comp. an instructive article by the author 
in Ztschr. d. gesamt. Staatswiss. for 1896, pp. i ff . : Der ofFentl. Haus- 
halt d. Stadt Fr. im Mittehlter.'-'ED.I 



2o8 THE DECLINE OF THE HANDICRAFTS, 

and the smith uses horseshoes purchased ready for 
use. But the customs connected with consumption change 
here but slowly; the demand remains more individual, and 
there is relatively far more repair work; indeed, the agri- 
cultural machines have brought fresh work of the latter 
type for iron-worker, smith, tinsmith, cooper, joiner. 
About fifty-two per cent, of the master craftsmen in Ger- 
many to-day are found in the country. The country has 
come to equal the cities in density of hand-worker popula- 
tion. Certainly the number of separate shops in the coun- 
try is particularly large. In Prussia the average num- 
ber of persons as assistants has seemingly diminished some- 
what since 1861; the number of apprentices is relatively 
high. But in this there is no ground for anxiety. The 
relation between the number of assistants and the number 
of masters is much more favourable to-day in rural parts 
than it was in the cities at the beginning of this century; 
and the condition of the rural craftsman, according to all 
that has been published on the subject, though modest, is 
still satisfactory. In this the reports to hand from 
Silesia, Saxony, East Friesland, Baden, and Alsace agree. 
There are certainly some village craftsmen leading very 
meagre lives; but such there have been in handicraft at 
all times. 

Among those who consider handicraft the ideal form 
of industrial activity two means have long been extolled 
for restoring solid footing and strength to the tottering 
industrial middle class; and there are many who still believe 
in their efficacy. 

The first is the "return to artistic work'' Efforts of this 
kind have been diligently fostered for well-nigh twenty- 
five years. For their encouragement museums, technical 
schools, and apprentice workshops have been instituted. 
But experience has soon taught, and the investigations of 



THE DECLINE OF THE HANDICRAFTS. 



209 



the Social Science Club have confirmed it anew, that these 
efforts have borne very little fruit for the small trader. 
Ironwork alone has gained at a few points through the re- 
newed employment of wrought-iron trellis-work, stair bal- 
ustrades, chandeliers, and the like. Otherwise all establish- 
ments successfully carrying on artistic industry are manu- 
facturing businesses of a large, and indeed of the largest, 
type. This is the case, for example, with bookbinding, art 
furniture, pottery. 

The second means is the extension of small power ma- 
chines and the electrical transmission of power, which shall 
enable the smallest master to obtain the most impor- 
tant labour-saving machines. Even men like Sir William 
Siemens and F. Reuleaux have placed the greatest hopes 
on the popularizing of these technical achievements. These 
expectations they have based upon the belief that success 
is simply a question of removing the technical superiority 
of the large undertaking, this superiority resting indeed 
in great part upon the employment of labour-saving ma- 
chines. 

In this they have curiously overlooked the fact that 
mechanical power is the more costly the smaller the scale 
on which it is employed. According to a table given by 
Riedel in the Centralblatt deutscher Ingenieure for 1891, 
the comparative expenses of a small motor working ten 
hours per day and horse-power are as follows (in cents — 
four pfennigs equal one cent) : 



Type of Motor.^ 



Small steam. ... 

Gas 

Compressed air. 

Electrical 

Petroleum 



Hone-power of Motor. 



1/4 



13 
16^ 



x/2 



9J 

7i 
I3f 

20 



7i 
6 

■n 

15 



a 



si 

4j 
5 

10 

8t 






' The price of gas is taken at 3 cents per cubic metre. 



4i 



6i 



31 

3\ 
4 



Si 



2IO THE DECUNE OF THE HANDICRAFTS. 

To place two businesses on a footing of technical equal- 
ity is thus not to g^ve them industrial equality. A machine 
must be fully utilized and able to pay for itself if it is to 
cheapen producton. As it cannot take over the whole 
process of production, but only individual parts of it, it 
presupposes, if it is to remain continuously in action, an 
expansion of the business, the employment of a larger 
number of workmen, greater outlays for raw material, 
rent of workshop, etc. For this the small master generally 
lacks the capital. Did he possess it, the advantages 
of more favourable purchase of raw material, of greater di- 
vision of work, of employment of the most capable tech- 
nical and artistic workmen, and of better chances of sale 
would always remain with the large undertaking.® It is 
difficult to imagine how shrewd men could overlook all 
Ihis. Has the tailor's, shoemaker's, or saddler's handicraft 
gained in vitality through the sewing-machine? 

The hope of finding through these two devices a new 
basis for handicraft must be abandoned; in most indus- 
trial branches in the larger cities there is no longer any 
such footing. Only in so far as the conditions of custom 

work continue unaltered does there remain room for a 
> 

limited number of businesses leavened with capital. In 
these other persons take the place of the craftsmen; small 
and moderately large entrepreneurs, foremen of the fac- 

• An interesting proof of what has been said is offered by the wood- 
turning machines in cabinet-making. None of the many larger handi- 
craft shops in the cabinet trade of Berlin (among which are also some 
well-founded businesses of moderate size, with twenty or more work- 
men) have adopted these machines in their work, although mechanical 
power of any strength is to be rented in many workshops in the city 
at a comparatively moderate price. It seems rather to be the case that 
small independent wage-paying shops have been opened which take 
charge of the cutting and fitting; and only the largest furniture fac- 
tories and cabinet-making establishments have set up those machines 
in their business. 



THE DECLINE OF THE HANDICRAFTS. 211 

tory workshops and skilled factory hands, contractors and 
home-workers. Externally all these groups, with the ex- 
ception of the last, are better situated than the majority 
of the small masters of the past. Whether they are better 
satisfied and happier is another question. 

Here, however, we are dealing rather with the tendency 
of the development than with the actual conditions of to- 
day. But we must not be deceived. The decline takes 
place slowly and silently; great misery, such as prevailed 
among the hand- weavers when they fought their forlorn 
battle against the mechanical spinning-mule, is found, with 
rare exceptions, only in the clothing industries. Certain 
grades of city population have ever remained true to the 
handicraftsman, and will be faithful for some time to come. 
There thus remains time for the coming generation to 
adapt itself to the new conditions. What it needs for the 
transition is a better general, mercantile and technical edu- 
cation. The thrifty, cautious person still finds opportunity 
to carry on work and gain a position; he is not so destitute 
and at a loss as those who leave school and workshop with 
insufficient equipment for life. 

It is our conviction that the process in question cannot 
be arrested by legislation, though it niay perhaps be re- 
tarded. But would that be a gain? 

In the preceding chapter the evolution of systems of 
industry was compared with the development of the ma- 
chinery of commerce, in which the earlier forms were, it 
is true, pressed back, though not destroyed, by the new. 
The comparison is applicable likewise to handicraft. 
Handicraft as a form of work is not perishing; it is only be- 
ing restricted to that sphere in which it can make the most 
of its peculiar advantages. That sphere to-day is the 
country, the districts where it still finds the conditions of' 
existence that gave birth to it in the Middle Ages. 



212 THE DECLINE OF THE HANDICRAFTS, 

In rural Germany we have at present, according to tol- 
erably exact estimates, some six hundred and seventy 
thousand master craftsmen and more than a half million 
journeymen and apprentices, or together about one and 
one-fifth millions engaged in active work. Adding the 
members of the masters* households we have, at a low 
estimate, a total of over three million persons. The largest 
part of this numerical success has been achieved by handi- 
craft in our own century. From the socio-political stand- 
point there is no ground for weeping with the masters of 
the small country hamlets who have lost their rural cus- 
tomers. Rather the contrary. 

During the period of the jealous exclusiveness of the 
town guilds, when one could pass on the highway thou- 
sands of journeymen who could nowhere obtain admit- 
tance to mastership, the journeymen smiths had a saying 
which the stranger at the meeting-house had to recite to 
the head, journeyman.® It ran: "A master I have not 
been as yet, but hope to become one in time, if not here, 
then elsewhere. A league from the ring, where the dogs 
leap and break the hedges [town limits], there it is good 
to be a master." 

Settlement in the country, at that time the sheet-anchor 
of the journeyman smith, still saves many thousands of 
craftsmen who do not feel themselves equal to the demands 
of city life. For the country an important social and eco- 
nomic advance lies in this admixture of industrial elements 
among the people; and the livelihoods there resting upon 
the foundation of handicraft are among the most whole- 
some offered by present society. Of course, they are to 
be measured by the natural standard of early hand-work. 



*Comp. L. Stock, Grundsuge d, Verfassung d, GeseUewwesens d. 
deutsch. Handwerketf p. 82. 



THE DECUNE OF THE HANDICRAFTS. 213 

not by the artificial standard taken from the phantasy of 
economic and political romancers. 

For just there lies the seat of the complaints and griev- 
ances which since the beginnings of modern development 
have been raised so persistently by the surviving repre- 
sentatives of urban craftsmen, that they have given rise to 
a false impression of the degree of comfort that handicraft 
as an industrial system can possibly afford its representa- 
tives. This standard was comparatively high in the Mid- 
dle Ages, because the hand-worker's position in life was 
then measured by the position of those in the social grade 
lying next below his own, from which he himself had often 
come — the class of villein peasantry, the " poor people " 
of the country. Compared with this unspeakably op- 
pressed class, handicraft had " golden soil," for it regu- 
larly yielded a money return and secured its members 
civil freedom, while the peasant was exposed to all the 
vicissitudes of agriculture and to the oppressions of the 
owners of the soil. It would be false to assume that the 
mediaeval craftsmen had on the average considerable cap- 
ital; and it was almost the same with the smaller traders. 
With what lay beyond — ^patrician families and nobility — 
the craftsman did not compare himself; under the system 
of classes founded upon birth the individual is satisfied if 
he obtains what is due his class. 

Our social system of to-day rests upon classes deter- 
mined by occupation. In such a system everyone com- 
pares himself with all others, because no legal barrier sep- 
arates him from the rest. In comparison with the other 
classes of modern society, the position of handicraft, 
even where it is still capable of holding its own, ap- 
pears a very modest one. All other classes would seem 
to have raised themselves, and the hand-worker class alone 
to have remained stationary. Where hand-work is strug- 



214 THE DECUNE OF THE HANDICRAFTS. 

gling for its very existence, it presents a sad picture of 
oppression. 

It is certainly not a spectacle to be viewed with com- 
posure to see that broad stratum of small independent 
persons who formed the heart of the early town popula- 
tions disappear and yield place to a disconnected mass of 
dependent labourers. It is a loss to society for which we 
find in urban soil no present compensation. 



CHAPTER VI. 

THE GENESIS OF JOURNALISM. 

The close connection existing in Germany between 
scientific investigation and university instruction, while 
exhibiting many unquestionably pleasing features, has this 
one great disadvantage, that those departments of knowl- 
edge which cannot form the basis of an academic career are 
inadequately investigated. This is the fate of journalism. 
While in France and England the history of journalism 
presents an extraordinarily rich and developed literature, 
we in Germany possess but two essays worthy of mention, 
one treating of the beginnings, the other, in a decidedly 
fragmentary manner, of the later development of the daily 
press.* 

In this condition of affairs there would be little profit 
in determining to which of the existing departments of 
scientific research this neglected task really falls. A sub- 
ject so complex as journalism can be treated with advan- 
tage from very different standpoints: from the standpoint 
of political history, of literary history, of bibliography, of 
law, of philology even, as writings on the slovenliness of 
journalistic style give proof. The subject is, without 
doubt, of most direct concern to the political economist. 

*Thc little book by Ludwig Salomon, Gesch. d. deutsch, Zeitungs- 
wesens von d, erst. Anfdng. bis 8. Wiederaufrichf, d. Deutsch. Retches, 
I, 1900, with its incomplete treatment of the subject, cannot materially 
alter this opinion. 

ai5 



1 



21 6 THE GENESIS OF JOURNALISM. 

For the newspaper is primarily a commercial contrivance, 
forming one of the' most important pillars of contemporary 
economic activity. But in vain do we search economic 
text-books, and even commercial manuals in a narrower 
sense, for a paragraph on the daily press. If, under these 
circumstances, we venture a brief and summary treatment 
of the beginnings of journalism, we are ourselves most 
fully conscious of our inability to make more than a partial 
presentation, and in so far as economic method is incapable 
of exhausting the material in all its phases, of the possible 
necessity of deceiving legitimate expectations. 

Our descriptions of the beginnings of journalism will 
vary with our conceptions of what a newspaper is. If the 
question. What is a newspaper? be put to ten different 
persons, perhaps ten different answers will be received. 
On the other hand, no one w^ho is asked to name the 
agencies that weave the great web of intellectual and ma- 
terial influences and counter-influences by which modern 
humanity is combined into the unity of society will need 
much reflection to give first rank to the newspaper, along 
with post, railroad, and telegraph. 

In fact, the newspaper forms a link in the chain of mod- 
ern commercial machinery; it is one of those contrivances 
by which in society the ex^change of intellectual and ma- 
terial goods is facilitated. Yet it is not an instrument of 
commercial intercourse in the sense of the post or the rail- 
way, both of which have to do with the transport of per- 
sons, goods, and news^ but rather in the sense of the letter 
and circular. These make the news capable of transport, 
only because they are enabled by the help of writing and 
printing to cut it adrift, as it were, from its originator, and 
give it corporeal independence. 

However great the difference between letter, circular, 
and newspaper may appear to-day, a little reflection shows 



THE GENESIS OF JOURNALISM, 21 J 

that all three are essentially similar products, originating 
in the necessity of communicating news and in the em- 
ployment of writing in its satisfaction. The sole difference 
consists in the letter being addressed to individuals, the 
circular to several specified persons, the newspaper to 
many unspecified persons. Or, in other words, while letter 
and circular are instruments for the private communication 
of news, the newspaper is an instrument for its publication. 

To-day we are, of course, accustomed to the regular 
printing of the newspaper and its periodical appearance 
at brief intervals. But neither of these is an essential char- 
acteristic of the newspaper as a means of news publication. 
On the contrary, it will become apparent directly that the 
primitive paper from which this mighty instrument of 
commercial intercourse is sprung appeared neither in 
printed form nor periodically, but that it closely resembled 
the letter from which, indeed, it can scarcely be distin- 
guished. To be sure, repeated appearance at brief inter- 
vals is involved in the very nature of news publication. 
For news has value only so long as it is fresh; and to pre- 
serve for it the charm of novelty its publication must follow 
in the footsteps of the events. We shall, however, soon 
see that the periodicity of these intervals, as far as it can 
be noticed in the infancy of journalism, depended upon the 
regular recurrence of opportunities to transport the news, 
and was in no way connected with the essential nature of 
the newspaper. 

The regular collection and despatch of news presupposes 
a wide-spread interest in public affairs, or an extensive 
area of trade exhibiting numerous commercial connec- 
tions and combinations of interest, or both at once. Such 
interest is not realized until people are united by some 
more or less extensive political organization into a certain 
community of life-interest. The city republics of ancient 



21 8 THE GENESIS OF JOURNALISM. 

times required no newspaper; all their needs of publica- 
tion could be met by the herald and by inscriptions as oc- 
casion demanded. Only when Roman supremacy had em- 
braced or subjected to its influence all the countries of the 
Mediterranean was there need of some means by which 
those members of the ruling class who had gone to the 
provinces as officials, tax-farmers and in other occupa- 
tions, might receive the current news of the capital. It is 
significant that Caesar, the creator of the military mon- 
archy and of the administrative centralization of Rome, is 
regarded as the founder of the first contrivance resembling 
a newspaper.^ 

We say resembling a newspaper, for journalism as now 
understood did not exist among the Romans; and Momm- 
sen's mention of a " Roman Intelligence Sheet " ' is but a 
distorted modernization. Caesar's innovations are to be 
compared rather with the bulletins and " laundry-lists " 
which the literary bureaux of our own governments supply 
for the use of journalists, than with our modern news- 
papers. Thus in his case it was not a question of founding 
journalism, but of influencing the newspapers already in 
existence. 

Indeed, long before Caesar's consulate it had become 
customary for Romans in the provinces to keep one or 
more correspondents at the capital to send them written 
reports on the course of political movement, and on other 
events of the day. Such a correspondent was generally an 

* Led ere, Des joumaux chez Us Romains (Paris, 1838). Lieberkuhn, 
De diurnis Romanorum actis (Vimar, 1840). A. Schmidt, Das Stoats^ 
zeitungswesen d. Romer, in Ztschr. L Geschichtsw., I, p. 303 ff. N. Zell, 
Ober d. Zeitungen d. alien Romer u, d, DodweWschen Fragmente, in his 
Ferienschriften, pp. i ff., 109 ff. Hubner, De senatus populique Romani 
actiSy in Fleckcisen's Jhrb. f. Philol., Suppl. Ill, pp. 564 ff. Heinze. 
De spuriis diumorum actorum fragmentis (Greifswald, i860). 

• Mommsen, Rom, Gesch,, III (4th ed.), p. 601. 



THE GENESIS OF JOURNALISM. 219 

intelligent slave or freedman intimately acquainted with 
affairs at the capital, who, moreover, often made a business 
of reporting for several. He was thus a species of primitive 
reporter, differing from those of to-day only in writing not 
for a newspaper, but directly for readers. On recom- 
mendation of their employers, tliese reporters enjoyed at 
times admission even to the senate discussions. Antony 
kept such a man, whose duty it was to report to him not 
merely on the senate's resolutions, but also on the speeches 
and votes of the senators. Cicero, when pro-consul, re- 
ceived through his friend, M. Caelius, the reports of a cer- 
tain Chrestus, but seems not to have been particularly well 
satisfied with the latter's accounts of gladiatorial sports, 
law-court proceedings, and the various pieces of city gos- 
sip. As in this case, such correspondence never extended 
beyond a rude relation of facts that required supplement- 
ing through letters from party friends of the absent per- 
son. These friends, as we know from Cicero, supplied the 
real report on political feeling. 

The innovation made by Caesar consisted in instituting 
the publication of a brief record of the transactions and 
resolutions of the senate, and in his causing to be pub- 
lished the transactions of the assemblies of the Plebs, as 
well as other important matters of public concern. 

The first were the Acta senatus, the latter the Acta diurna 
populi Romanu The publication was made by painting the 
text on a white tablet smeared with gypsum. The tablet 
was displayed publicly, and for the inhabitants of the cap- 
ital was thus what we call a placard. For those abroad 
copies were made by numerous writers and forwarded to 
their employers. After a certain interval the original was 
placed in the archives of the state. 

This Roman Public Bulletin was thus not in itself a 
newspaper, though it attained the importance of such by 



220 THE GENESIS OF JOURNAUSM. 

what we would consider the cumbersome device of private 
correspondence to the provinces. 

The Acta senatus were published for but a short time, 
being suppressd by Augustus. On the other hand, the 
Acta diuma populi Romani obtained such favour in the eyes 
of the people that their contents could be made much more 
comprehensive, while their publication was long con- 
tinued under the Empire. They more and more developed, 
however, into a kind of court circular, and their contents 
began to resemble the matter offered by the official or 
semi-official sheets of many European capitals to-day. On 
the whole, they confined themselves to imparting facts; 
their one noticeable tendency was to ignore the disagree- 
able. 

The contents still continued to reach the provinces by 
way of correspondence; and, as Tacitus tells us, the people 
had regard not merely for what the official gazette con- 
tained, but also for what it left unrecorded: people read 
between the lines. How long the whole system lasted we 
do not know. Probably after the removal of the court to 
Constantinople it gradually came to an end. 

The Germanic peoples who, after the Romans, assumed 
the lead in the history of Europe, were neither in civiliza- 
tion nor in political organization fitted to maintain a sim- 
ilar constitution of the news service; nor did they require 
it. All through the Middle Ages the political and social 
life of men was bounded by a narrow horizon; culture re- 
tired to the cloisters, and for centuries affected only the 
people of prominence. There were no trade interests be- 
yond the narrow walls of their own town or manor to draw 
men together. It is only in the later centuries of the 
Middle Ages that'' extensive social combinations once more 
appear. It is first the church, embracing with her hierarchy 
all the countries of Germanic and Latin civilization, next 



THE GENESIS OF JOURNALISM. 221 

the burgher class with its city confederacies and common 
trade interests, and, finally, as a counter-influence to these, 
the secular territorial powers, who succeed in gradually 
realizing some form of union. In the twelfth and thir- 
teenth centuries we notice the first traces of an organized 
service for transmission of news and letters in the messen- 
gers of monasteries, the universities, and the various spirit- 
ual dignitaries; in the fourteenth and fifteenth centuries we 
have advanced to a comprehensive, almost postlike, organ- 
ization of local messenger bureaux for the epistolary inter- 
course of traders and of municipal authorities. And now, 
for the first time, we meet with the word Zeitung, or news- 
paper. The word meant originally that which was happen- 
ing at the time (Zeit = time), a present occurrence; then 
information on such an event, a message, a report, news. 

In particular do we find the word used for the com- 
munications on current political events which were re- 
ceived by the town clerks from other towns or from indi- 
vidual friends in the councils of those towns, either as let- 
ters or as supplements to them, and which are still fre- 
quently found in their archives. Thus the municipal 
archives of Frankfurt-on-Main possess as many as 188 let- 
ters relating to the raids of the Armagnacs in the early 
forties of the fifteenth century; they are mostly descrip- 
tions of sufferings and appeals for help from towns in Al- 
sace and Switzerland. Among them are not less than 
three accounts of the battle of St. Jacob, one from Zurich, 
one from Strassburg, and one from the council of Basel.* 

*Wulker, Urkunden u. Schreiben, hetreff. d. Zug d. Armagndken: in 
Neujahrsblatt d. Vereins f. Gesch. u. Altertumsk. zu Frankfurt-a.-M. 
for 1873. 

On the following section consult: Hatin, Histoire politique et littiraire 
de la presse en France (Paris, 1859-61), Vol. I, pp. 28flF., and his Biblio- 
graphie historique et critique de la presse pSriodique frangaise, precidSe 
d^un Essai historique et statistique sur la naissance et les progrh de la 



222 THE GENESIS OF JOURNAUSM. 

This reporting is voluntary, and rests upon a basis of 
reciprocity. It sprang from the common interest uniting 
the towns against the noble and the territorial powers, 
and found effective support in the numerous town- 
messengers who maintained in regular courses — for this 
reason called " ordinary " messengers — ^the connection 
between Upper and Lower Germany. 

In the fifteenth century we find a similar exchange of 
news by letter between people of high standing — ^princes, 
statesmen, university professors — which reaches its high- 
est development during the era of the Reformation. It is 
now good form to add to a letter a special rubric, or to 
insert on special sheets, " Novissfima'' " Tidings," " New 
Tidings " or " News," *' Advices." Moreover, we already 
notice how people have ceased to give each other mere 
casual information about the troubles and distress of the 
time, and aim at a systematic collection of news. It was 
especially to the g^eat commercial centres and the trading 
towns which were the centres of messenger activity and 
the seat of higher education that news items flowed from 
all quarters, there to be collected and re-edited into letters 
and supplements, and thence to be diverted in streams in 
all directions. Everywhere these written tidings bear the 
name of newspaper (Zeitung or neue Zeitungen), 

The largest part of this correspondence is of a private 



presse piriodique dans Us Deux Mondes (Paris, 1866), pp. xlviiff.; 
Leber, De Tetat rSel de la presse et des pamphlets depuis Francois I 
jusqu'a Louis XIV (Paris, 1834); Alex. Andrews, The History of 
British Journalism (London, 1859), Vol. I, pp. 12 ff.; Ottino, La stampa 
periodica, il commercio dei lihri e latipographia in Italia (Milano, 1875), 
p. 7; Rob. Prutz, Geschichte d. deutsch. Joumalismus (Hanover, 1845), 
VoL I: J. Winckler, Die period, Presse Oesterreichs (Vienna, 1845), pp. 
19 ff.; Grasshoff, Die bricHiche Zeitung d. XV L Jahrhunderts (Leipzig, 
1877): Steinhausen, in Archiv f. Post u. Telegraphic, 1895, pp. 347 ff., 
and his Geschichte d. deutsch. Brief es, 2 vols. 



THE GENESIS OFJOURNAUSM. 223 

character. Men at the centre of political and ecclesiastical 
activity communicated to each other the news that had 
come to hand. It was a reciprocal giving and receiving 
which did not prevent those with a heavy correspondence 
from multiplying their news-sheets in order to append 
them to letters to different persons, nor the recipients 
from redespatching copies of them or circulating them 
amongst their acquaintances. Princes, it would seem, al- 
ready maintained at important commercial points their 
own paid correspondents. 

For a time these written newspapers did not find their 
way among the masses. The circles for which they were 
intended were: (i) princes and statesmen, as also town- 
councillors; (2) university instructors and their immediate 
cooperators in the public service in school and church; 
(3) the financiers of the time, the g^eat merchants. 

Almost all reformers and humanists are diligent news- 
paper correspondents and regular recipients of newspaper 
reports. This is especially true of Melancthon, whose 
numerous connections throughout all parts of Germany 
and the neighbouring countries continually brought him a 
plentiful store of fresh news, with which he in turn sup- 
plied his friends, and certain princes in particular. In 
comparison with his, Luther's and Zwingli's correspond- 
ence is relatively poor in such matter. On the other hand, 
John and Jacob Sturm, Bucer and Capito of Strassburg, 
Oecolampadius and Beatus Rhenanus of Basel, Hatzer and 
Urbanus Rhegius of Augsburg, Hier. Baumgartner of 
Nuremberg, Joachim Camerarius, Bugenhagen, and oth- 
ers were very zealous and active in this direction. 

The sources for their news are manifold. Besides oral 
or written communications from friends, we know of nar- 
ratives of incoming merchants, particularly of book-dealers 
who had visited the Frankfurt fair, reports of letter-car- 



a 24 THE GENESIS OF JOURNAUSM. 

riers, accounts from soldiers returning home from their 
campaigns, communications from strangers passing 
through their town or from visiting friends, and especially 
from students coming from foreign lands to study at Ger- 
man universities; finally, any items gleaned from foreign 
ambassadors who happen to be passing through; from 
chancellors, secretaries, and agents of important person- 
ages. 

Naturally such oral news collected at random varied 
greatly in worth, and had first to undergo the editorial 
criticism of the correspondent before being circulated 
further. The news-items based upon written information 
were of much greater importance. It may be of some 
interest, by following Melancthon's correspondence, to in- 
quire somewhat into the sources of them.* 

We soon perceive that there were a number of definite 
collecting centres for the various classes of news. In the 
forefront of interest at that time stood the Eastern ques- 
tion, that is, the threatening of the countries of Central • 
Europe by the Turks. News of the engagements with the 
latter came either from Hungary through Vienna, Cra- 
cow or Breslau, or from Constantinople by sea by way of 
Venice. The reporters are mostly ecclesiastics, adherents 
of the New Learning. 

On affairs in the South communications came from 
Rome, Venice, and Genoa, as well as from learned friends 
in Padua and Bologna. News from France and Spain was 
procured by way of Lyons, Genoa, and Strassburg; from 
England and the Netherlands by way of Antwerp and Co- 
logne; from the countries of the North by way of Bremen, 
Hamburg, and Lubeck; from the Northeast by way of 
Konigsberg and Riga. 

' According to Grasshoff, cited above, pp. 23 ff. 



THE GENESIS OF JOURNALISM, 225 

In Germany, Nuremberg was the chief collecting centre 
for news, partly by reason of its central position, partly 
because of its extensive trade connections. Anyone de- 
sirous of receiving reliable and definite information on the 
doings of the world wrote to Nuremberg or sent thither 
a representative. Princes like Duke Albert of Prussia 
and Christian III of Denmark there maintained resident 
correspondents, whose duty it was to collect and report 
any fresh items of news. Town officials, councillors, and 
reputable merchants frequently undertook such an office. 
Besides Nuremberg, Frankfurt, Augsburg, Regensburg, 
Worms, and Speier were also important news centres. 

The newspapers that Melancthon compiled from these 
various sources are simply historical memoranda, selected 
with some care and interspersed at rare intervals with 
discussions of a political nature, and more frequently 
with all kinds of complaints and fears, wishes and hopes. 
Along with the important news from the Emperor's court, 
from the various seats of war and on the progress of the 
Reformation, we meet with others reflecting the complete 
naivete and incredulity of the times: reports of political 
prophecies, strange natural phenomena, missbirths, earth- 
quakes, showers of blood, comets and other celestial ap- 
paritions. 

In the second half of the sixteenth century this species 
of news-agency received definite form and organization as 
a business, not merely in Germany, but, apparently even 
earlier, in Italy, especially in Venice and Rome. 

Venice was long regarded as the birthplace of the news- 
paper in the modern acceptation of the word. . This opin- 
ion was supported by the . extensive use of the name ga- 
jsetta or gazette amongst the Latin nations for a newspaper; 
and this word is to be found earliest in Venice as the name 
of the small coin. 



226 THE GENESIS OF JOURNAUSM. 

We will not enter into the accounts — ^at times rather 
romantic — that have been given to justify the derivation 
(in itself improbable) of the name of the newspaper from 
the name of the coin.* 

In itself, however, there is much to be said for the pre- 
sumption that journalism, as described above, was first de- 
veloped as a business in Venice. As the channel of trade 
between the East and West, as the seat of a govern- 
ment that first organized the political news service and 
the consular system in the modern sense, the old city of 
lagoons formed a natural collecting centre for important 
news-items from all lands of the known world. Even 
early in the fifteenth century, as has been shown by the 
investigations of Valentinelli, the librarian of St. Mark's 
Library, collections of news had been made at the instance 
of the council of Venice regarding events that had 
either occurred within the republic or been reported by 
ambassadors, consuls, and officials, by ships' captains, mer- 
chants and the like. These were sent as circular despatches 
to the Venetian representatives abroad to keep them 
posted on international affairs. Such collections of news 
were called fogli d'awisi. 

Later on, duplicates of these official collections were 
made, though evidently not for public circulation, but 
rather for the use of prominent citizens of Venice who 
sought to derive advantage from them in their commercial 
operations, and also communicated them by letter to their 
business friends in other lands. 

This appending of political news to commercial corre- 
spondence, or the enclosing of the same on special sheets, 
soon became the practice also among the large traders 
of Augsburg, Nuremburg, and the other German towns. 

* Comp. Hatin, Bibliographie de la presse piriodique, p. xlviL 



THE GENESIS OF JOURNALISM, 227 

By and by it occurred to some that the collection and 
transmission of news by letter could be made a source of 
profit. In the sixteenth century we find on the Venetian 
Rialto, between the booths of the changers and gold- 
smiths, an independent news-bureau that made a business 
of gathering and distributing to interested parties polit- 
ical and trade news: information as to arrival and clearance 
of vessels, on prices of wares, on the safety of the high- 
ways, and also on political events.*^ Indeed, a whole guild 
of scrittori d'awisi grew up. In a short time we meet with 
the same people in Rome, where they bear the name nov- 
ellanti or gazettanti. Here their activity, whether because 
they circulated disagreeable facts or accompanied their 
facts with their own comments, became discomforting to 
the Curia. In the year 1572 not less than two papal bulls 
were issued against them (by Pius V and Gregory XIII); 
the writing of " advices " was strictly forbidden, and its 
continuance threatened with branding and the galleys. 
Nevertheless we continue to meet numerous indications 
of a news service from Rome to the Upper Italian cities 
and to Germany. 

In the meantime, newspaper writing had also become a 
business in Germany with an organization that, for the 
existing conditions of trade, is really wonderful. This 
organization is connected on the one hand with the further 
development of despatch by courier, and on the other with 
Emperor Maximilian's institution of the post from the 
Austrian Netherlands to the capital, Vienna, by which the 
regular receipt of news was greatly facilitated. We thus 
find, at various places, in the second half of the sixteenth 
century, special correspondence bureaux which collect and 
communicate news by letter to their subscribers. Several 

' According to Pnitz. Gesch. d, Joumalismus, I, p. 212. 



228 ^ THE GENESIS OF JOURNALISM. 

collections of these epistolary newspapers have been pre- 
served: for instance, one from 1582 to 1591 in the Grand 
Ducal library in Weimar, and two in the University li- 
brary at Leipzig from the two last decades of the same 
century.® 

Let us refer briefly to the oldest year of the Leipzig col- 
lection. It bears the heading: " News to hand from 
Nuremberg from the 26th of October Anno '87 to the 26th 
of October Anno '88." Then follow in independent group- 
ings transcripts of the news received weekly from Rome, 
Venice, Antwerp, and Cologne at the office of the Nurem- 
berg firm of merchants, Reiner Volckhardt and Florian 
von der Bruckh, and thence given out again either by the 
house itself or by a special publisher. The person who re- 
ceived the present collection was probably the chief city 
clerk of Leipzig, Ludwig Triib. 

The communications from Rome are as a rule dated 
about six days earlier than those from Venice, and the 
Antwerp correspondence about five days earlier than that 
from Cologne. All four places lay on the great post-routes 
from Italy and the Netherlands to Germany. Along with 
these periodical communications irregular ones appear 
now and then, for instance, from Prague and Breslau, and 
particularly often from Frankfurt-on-Main. 

Examining the contents of these news-items more 
closely we soon find that we have to do not with events 
occurring in Rome, Venice, Antwerp, etc., but with re- 
ports collected at these places. Thus the correspondence 
from Antwerp contained not merely news from the Neth- 
erlands, but also from France, England, and Denmark; by 
way of Rome came news not only from Italy, but from 
Spain and the south of France as well; from Venice came 

• Comp. Jul. Opel, Die Anfdnge d, deutsch, Zeitungspresse in Archiv 
f. Gesch. d. deutsch. Buchhandels, Vol. Ill (1859). 



THE GENESIS OF JOURNALISM, 229 

news from the Orient. The reports are soberly descriptive 
and commercial in tone. Political items preponderate; 
communications on trade and commerce appear less fre- 
quently. There is no trace of the favourite tales of won- 
ders and ghosts. 

But how was the news service in these four great collect- 
ing points organized? Who were the collectors and the 
intermediaries? How were they paid? What were their 
sources of information? Unfortunately we can answer 
only part of these questions. 

In the first place, as to the sources from which the au- 
thors of the letters derive their information, they them- 
selves appeal at times even to the last mail or to the regular 
messenger service, the " Ordinari," Thus we read in a 
letter from Cologne dated February 28, 1591: "The let- 
ters from Holland and Zeeland, and also from Italy, have 
not yet appeared." In a similar letter from Rome of date 
February 17, 1590, we are informed that the postmaster 
there has contracted with the Pope to establish a weekly 
post to and from Lyons; and at the close we read, " In this 
way we shall have news from France every week." 

Nothing more than this is to be gleaned from the collec- 
tion itself. When, however, we notice contemporaneously 
in several German cities that it is the heads of the town- 
couriers and the imperial postmasters who in particular de- 
vote themselves to the business of editing and despatching 
news-letters, the supposition gains greatly in probability 
that the collection of news is in most intimate connection 
with the mail service of the time. The messenger mas- 
ters and the postmasters probably exchanged at regular 
intervals the news they had collected, in order to pass it 
on to their particular clients. But the whole matter 
stands greatly in need of closer investigation.^ 

• Steinhausen in Archiv f. Post u. Tel., 1895, p. 355, expresses merely 



230 THE GENESIS OF JOURNALISM, 

The relations between wholesale trade and newspapers 
are somewhat clearer. Like the Nuremberg merchants 
mentioned above, some large trading houses in other local- 
ities had also organized ah independent news service. Es- 
pecially prominent were the Welsers and Fuggers, whose 
news reports we find in the celebrated letter-book of 
the Nuremberg jurist, Christoph Scheurl,^^ along with the 
Nuremberg correspondence. In the second half of the six- 
teenth century the Fuggers had the news coming to them 
from all parts of the world, regularly collected and appar- 
ently also published. The title of the regular numbers was 
" Ordinari'Zeittungeny There were also supplements, or 
" specials," with the latest items. The price of one num- 
ber was four kreuzer; the yearly cost in Augsburg, includ- 
ing delivery, was 25 florins, and for the ordinari numbers 
alone, 14 florins. One Jeremiah Krasser, of Augsburg, 
burgher and newspaper writer, is named as editor. He 
informs us that he supplied many other gentlemen in 
Augsburg and district with his news. A file of this organ 
of publication, so rich in material, for the years 1568 to 
1604 is found in the Vienna library.^ ^ 

The newspapers of the Fuggers. regularly contain news 
from all parts of Europe and the East, and also from places 
still further removed: Persia, China and Japan, America. 
Besides the political correspondence, we have frequent re- 
ports of harvests and memoranda of prices, now and then 
even communications in the nature of advertisements, and 
a long list of Vienna firms — how and where all things 
could now be procured in Vienna. Even literary notices 

a supposition, though indeed a very well grounded and probable one, 
on the course of development. 

"Christoph Scheurl: Brief buck, ein Beitrag s. Geschich. d. Reforma- 
Hon u. ihrer Zeit (Sooden u. Knaake, Potsdam, 1867-1872). 

" Sickel, Weimar. Jahrb. f. deutsche Sprache u. Litteratur, I, p. 346, 



THE GENESIS OF JOURNALISM. 231 

of recent and noteworthy books appear; and there is one 
account of the presentation of a new drama. 

As in Augsburg, so in other places in Germany we meet 
individual correspondents — ^journalists (Zeitunger), novel- 
ists — ^who carry on their newspaper writing in the service 
of princes or of cities. Thus in 1609 the elector of Saxony, 
Christian II, made a contract with Joh. Rudolf Ehinger, 
of Balzhein in Ulm, whereby the latter undertook for a 
yearly fee of 100 florins to furnish reports upon events in 
Switzerland and France, Swabia naturally being included. 
In the year 161 3 Hans Zeidler, of Prague, received from 
the Saxon court for similar service a yearly salary of 300 
florins, together with 3319 thalers 6 g. gr. for expenses 
incurred in collecting his news.*^ In the same year the 
sovereign bishop of Bamberg had newspapers forwarded 
to him by a Dr. Gugel, of Nuremberg, for a fee of 20 flor- 
ins. In the year 1625 the town of Halle paid the news cor- 
respondent, Hieronymus Teuthorn, of Leipzig, the sum of 
two schock, eight groschen, as quarterly fee; and as late as 
1662 the council of the town of Delitzsch was subscriber to 
a newspaper correspondence from Leipzig at a quarterly 
fee of two thalers. The postmasters and messenger chiefs 
appear to have been paid somewhat better for their ser- 
vices, which were indeed more valuable. At least, we 
know that in the year 161 5 the postmaster at Frankfurt, 
Johann von der Birghden, who furnished a great number 
of German princes with news,^' was in receipt of a yearly 
salary of 60 florins for supplying the electoral court of 
Mainz with the weekly newspapers.^* 

"C. D. V. Witzlcben, Gesch. d. Leipsiger Zeitung (Leipzig, i860), ppt 
5-6. The Saxon court in 1629 maintained similar agents in Vienna, 
Berlin, Brunswick, Augsburg, Ulm, Breslau, Hamburg, Lubec, 
Prague, Amsterdam, at the 'Hague, and in Hungary. 

" Comp. Opel, as above, pp. 28, 66. 

" Faulhaber, Gesch, d. Post in Frank furt-a.-M. (Archiv. f. Frankf. 
Gesch. u. Kunst, New Series, X), pp. 31, 60 fF. 



232 THE GENESIS OF JOURNALISM. 

Even in the seventeenth century the written newspapers 
appear not to have made their way to wider circles. They 
were still too costly for that. 

At the close of the sixteenth and in the seventeenth 
century we find written newspapers in France and Eng- 
land, as well as in Germany and Italy. In France they are 
called nouvelles d la main; in England, news-letters. In both 
countries they are confined to the capital city. 

The line of development in Paris is the more interest- 
ing: it may be said indeed that the most primitive of all 
newspapers, the precursor of the written newspaper, is to 
be found there. It is the related, or spoken, paper.^** 

In the turbulent times of the sixteenth and seventeenth 
centuries groups of Parisian burghers would assemble each 
evening on the street corners, on the Pont Neuf, and on 
the public squares, bringing together the news of the day 
and making their own comments upon it. As is easily 
conceivable, there were among these groups individuals 
who became adepts in the collection and repetition of news. 
Gradually method and organization were introduced; the 
so-called nouvellistes held regular meetings, exchanged 
their news with each other, and made comments thereon, 
discussed politics, and laid plans. 

" Comp. Hatin, Histoire de la presse en France, Vol. I, pp. 32-33. [An 
interesting present-day instance of the " spoken " newspaper, which 
may indeed not be so very rare a phenomenon, is given in a sketch 
of Swiss life in the little village of Champery by a recent writer in 
the Canadian Magazine. " On three of the houses of the village/' it 
is stated, *' are curious balconies, which are in reality old pulpits, 
once used for open-air preaching. They now serve the place of the 
country newspaper, for on Sundays, after mass, a man calls out from 
them the news of the week, what there is for sale, what cattle have 
been stolen or have strayed, and other items of interest to those who 
have come down for the day from the isolation of the high moun- 
tains." — Swiss Life and Scenery, by E. Fanny Jones. Can. Mag., Aug. 
1898, p. 287.— Ed.] 



THE GENESIS OF JOURNALISM. 233 

The writers of the time treat these groups with never- 
ending satire; the comic dramatists seize the fruitful theme, 
and even Montesquieu devotes to them one of the most 
entertaining of his Lettres Persanes}^ 

What was at the outset a mere pastime for news-hunters 
and idlers, enterprising brains soon developed into a busi- 
ness. They undertook to supply .regular news to people 
of rank and standing. Men in high station kept a nouvel- 
liste as they kept a hair-dresser or surgeon. Mazarin, for 
instance, paid such a servant 10 livres per month. 

These groups of nouvellistes soon began to seek cus- 
tomers in the Provinces also, and these, of course, could be 
supplied only by letter. Each group had its particular 
editorial and copying bureau, and its special sources for 
court and official news. The subscribers paid a fixed sum, 
according to the number of pages that they desired each 
week. Thus originated the celebrated nouvelles d la main, 
which, in spite of many prosecutions on the part of the 
government, lasted till well towards the end of last cen- 
tury, and which were often sent abroad as well.^^ That 
which gave them a firm footing along with the printed 
newspapers was, to a certain extent, the circumstance that 
they rendered the secrecy of the government system 
largely illusory, and further, took the liberty rioW and 
then of animadverting on public conditions.^® 

In England likewise the news-letters, more especially 
devoted to furnishing the country nobility with the news 
of the capital and the court, maintained themselves well 

"(Euvres computes (Paris, 1857), p. 87, Lettre CXXX. 

^^La Gazette de la Rigencey Janvier 171$ — juin 17 19, publiie d'aprks U 
manuscrxt inidU conservi d la Biblioth^que royale de La Haye, par Le Comte 
E, de Barthilemy (Paris, 1887) gives a description of the contents of 
these sheets. 

*• Similarly in Austria: Joh. Winckler, Die period. Presse Oesterreichs 
(Vienna, 1875), pp. 28-9. 



234 THE GENESIS OF JOURNAUSM. 

into the eighteenth century. The printed newspapers of 
that epoch indeed adapted themselves to the system to 
the extent of appearing with two printed and two un- 
printed pages, thereby enabling their subscribers to send 
them to others, enriched with additional notes in writing.^® 

Thus we see that at about the same time in all the ad- 
vanced countries of Europe the written newspaper arises 
as a medium — of course, as yet a decidedly restricted me- 
dium — of news publication, and maintains itself for more 
than two hundred years. It is, however, most remarkable 
that the production of these written news-sheets as a busi- 
ness can nowhere be traced beyond the period of the in- 
vention of printing. In this connection the question 
naturally arises, why the printing-press was not taken into 
the service of the regular news publication. 

The question is answered by the simple fact that even in 
young colonial countries with an European population ac- 
customed at home to printed newspapers, the written pre- 
ceded the printed news-sheets. This was true of the 
English colonies in America at the beginning of the eigh- 
teenth century ,^^ and of the colony of West Australia in 
1830.^^ This proves that it could not have been so much 
the pressure of the censorship, that so long delayed the 
employment of the press for news publication, as the lack 



19 



For details: Andrews, The History of British Journalism, Vol. I, 
pp. 14 ff.; Hatin, as above, p. 51. Joachim von Schwarzkopf, Uehcr 
Zciiungen (Frankfurt-on-Main, 1795), p. 9, relates that likewise in 
Germany " in the case of some newspapers that in contents and form 
resembled manuscript sheets (for example in Mainz and Regensburg) 
the printing-press was occasionally made use of because of the large 
number of subscribers." He also mentions Vienna, Munich, Berlin 
and Hanover as places from which sheets filled with secret domestic 
news were distributed. 

^ Frederic Hudson, Journalism in the United States from j6^ to 1830 
(New York, 1873), pp. 51 ff. 

** Andrews, as above, Vol. II, pp. 312 ff. 



THE GENESIS OF JOURNALISM. 235 

of a sufficiently large circle of readers to guarantee the 
sale necessary to meet the cost of printing. 

However, since as far back as the close of the fifteenth 
century, special numbers of those written newspapers con- 
taining matter of a presumably broader interest were fre- 
quently printed. These were the one-page prints issued by 
enterprising publishers under the name of " Neue Zeitung/' 
and disposed of at fairs and markets. Collections are to 
be found in every old library.^^ The oldest of these 
prints is a report of the obsequies of Emperor Frederick 
III from the year 1493. From that time till the sixteenth 
century had run its course they continued to hold their 
own; but with the growth of periodical news-sheets, 
they became rarer, and finally, in the eighteenth, disap- 
peared. The earliest numbers bear either no title at all or 
take their title from the contents. The name Zeitungy 
or newspaper, for such a loose sheet appears for the first 
time in 1505. We find, however, various other appella- 
tions; for example, Letter, Relation, Tale, News, Descrip- 
tion, Report, Advice, Post, Postilion, Courier, Rumour, 
Despatch, Letter-bag.^® To these all kinds of qualifying 
titles are frequently added, such as Circumstantial News, 
Truthful and Reliable Description, Faithful Description, 
Truthful Relation, Review and Contents, Historical Dis- 
course, and Detailed Explanation; very often we have: 
New and Truthful Tidings, Truthful and terrible Tidings, 
Wonderful, terrible, pitiful Tidings. In England some of 
the titles are: Newes, Newe Newes, Thiding, Woful 
Newes, Wonderful and strange News, Lamentable News; 

" Treated- bibliographically by Weller, Die ersten deutsch. Ziitungen 
(Bibliothek d. literar. Vereins, Vol. LXI). Supplement to same in 
the " Germania," XXVI, p. 106. 

^ Brief, Relation, Mar, Nachricht, Beschretbung, Berichf, Aviso, Post, 
PosHUion, Kurier, Fatna, Depesche, FeUeisen. 



9$6 THE GENESIS OF JOURNAUSM. 

and in France: Discours, Memorable Discours, Nouvelles, 
Recit, Courier, Messager, Postilion, Mercure, etc. 

The titles, we notice, are sensational and pretentious. 
The contents vary greatly. In the gr^at majority of cases 
they consist of political news; argument remains altogether 
in the background. The written news-letters are the chief, 
though not the sole, source for these fugitive productions 
of the printing-press. Ordinarily, the one-page prints are 
independent of each other; and only here and there at the 
end of the sixteenth century can several consecutive num- 
bers be instanced; but this is not sufficient to justify us in 
supposing a periodical issue. These loose sheets, however, 
at least prepare the way, as regards form and matter, for 
the printed newspaper with its regular issues. And they 
render a like service in so far as they awakened among the 
masses an interest in occurrences reaching beyond mere 
parish affairs. 

The first printed periodical news collections begin as 
early as the sixteenth century. They are annual publica- 
tions, the so-called Postreuter (postilions) or news epit- 
omes whose contents may in a manner be compared with 
the poHtical reviews of the year in our popular calendars.** 

These are supplemented by semi-annual news sum- 
maries, the so-called Relationes semestrales or Fair Reports. 
They were begun between 1580 and 1590 by Michael von 
Aitzing. They drew their information chiefly from the 
regular post and traders' newspapers, and for more than 
two centuries formed one of the chief articles of sale at 
the Frankfurt fair, and later on at the Leipzig spring and 
fall fairs as well.*' The first printed weekly of which we 

•• According to Prtitz, cited above, p. 179, they appeared as early as 
the middle of the sixteenth century. 

"F. Stieve, Ueber d. altest. kalbj'dhrig. Zeitungen oder Messrelationen, 
u, insbesond, uber deren Begriinder. Frhm. Michael von Aitzing: Abh. 



THE GENESIS OF JOURNALISM, 237 

have direct information is a Strassburg sheet, whose num- 
bers for the year 1609 are found in the University library 
at Heidelberg, while fragments of later years are pre- 
served in the public library at Ziirich.^* It corresponds 
exactly in matter and form with the regular despatches 
which the post brought weekly from the chief collecting 
places of the news trade. It was soon imitated; and after 
the beginning of the Thirty Years' War the growth in the 
number of printed weeklies was particularly rapid. We 
have evidence of the existence of about two dozen in the 
second and third decades of the seventeenth century. They 
were established chiefly by book-printers; though in 
numerous places the post assumed, naturally with varying 
success, the right of printing despatches as a part of its 
prerogative. While in Frankfurt, Leipzig, Munich, Col- 
logne, and Hamburg the old connection between post and 
newspaper continued for a considerable time, the publica- 
tion of news was in many other towns completely absorbed 
by the book-printers, a fact of the greatest moment for its 
further development. 

Germany is the first country that can show printed 
newspapers appearing regularly at brief intervals. The 
English and the Dutch claims to the honour of having 
produced the earliest printed weeklies are now generally 
abandoned. England can point to nothing similar before 
the year 1622: the first French weekly sheet appeared in 
1631.27 

der k. baycr. Akad. d. Wiss., Ill, G. XVI, p. i (Munich, 1881). 
Comp. also Orth, AtufuhrL Abhand. von d, beriihmten swoen Reichs- 
messetiy so in d. Reichsstadt Frank furt-a.-M, jdhrlich abgehalten werden 
(Frankfurt, 1765), pp. 714 ff.; Prutz, as above, pp. 188 ff.; J. von 
Schwarzkopf, Ueber potitische u. gelehrte Zeitungen in Frank furt-a.-M., 
1802. 

" Opel, as above, pp. 44 ff. 

" [See article Newspapers in the Ency. Brit, and the literature there 
given. — Ed.] 



2^S THE GENESIS OF JOURNALISM. 

It will perhaps seem strange that a leap was made di- 
rectly from half-yearly reports to weekly publications 
without a transitional stage of monthly reports. It must, 
however, not be forgotten that the collection of news, as 
well as the distribution of the news-sheets, had to conform 
to the peculiar cpmmercial facilities of the time. The most 
important of these were the fairs and the stage-posts. The 
semi-annual fairs made possible the distribution of the 
printed news from one great centre of trade and travel to 
even the most remote points. But the stage-posts tra- 
versed the chief trade-routes once a week each way. The 
leap from the half-yearly to the weekly reports lay thus in 
the nature of things. 

By the weekly newspapers the impetus was given to the 
essentially modern development of the press. Yet it was a 
considerable time before the first daily newspapers ap- 
peared. This occurred in Germany in 1660 {Leipziger 
Zeitung), in England in 1702 (Daily Courant), in France in 
1777 (Journal de Paris). 

We need not pursue this theme down to our cosmopol- 
itan papers that appear three times a day. The distin- 
guishing feature of the latter as contrasted with the written 
newspaper of the sixteenth century is not so much the 
magnitude of the organization for procuring news and the 
rapidity in transmitting it, as the transformation in the 
nature of the contents, particularly the advertising, and 
the influence thereby exerted on public opinion, and, con- 
sequently, on the course of the world's history. 

For the sixteenth century the network of agencies for 
the regular collection of news already described, was with- 
out doubt magnificent. There runs through it a modern 
characteristic, the characteristic of uniting individual 
forces in divided labour towards a single end. In the de- 
partment of news collection there has been little advance 



THE GENESIS OF JOURNAUSM. 239 

since the sixteenth century. The whole subsequent de- 
velopment of the newspaper in this direction rests on the 
separation of news collection from news despatch (post), 
and on the commercial organization of the former into 
correspondence bureaux and telegraph ag'encies. To the 
telegraph agencies have fallen the duties of the earlier 
postmasters and news-scribes, but with this difference, that 
they no longer labour directly for the newspaper readers, 
but supply the publishing house with half-finished wares, 
making use in such work of the perfected commercial ma- 
chinery of modem times. 

Again, the further development of news publication in 
the field that it has occupied since the more general 
adoption of the printing-press, has been peculiar. At the 
outset the publisher of a periodical printed newspaper dif- 
fered in no wise from the publisher of any other printed 
work — for instance, of a pamphlet or a book. He was but 
the multiplier and seller of a literary product, over whose 
content he had no control. The newspaper publisher mar- 
keted the regular post-news in its printed form just as an- 
other publisher offered the public a herbal or an edition of 
an old writer. 

But this soon changed. It was readily perceived that 
the contents of a newspaper number did not form an 
entity in the same sense as the contents of a book or pam- 
phlet. The news-items there brought together, taken 
from different sources, were of varying reliability. They 
needed to be used judicially and critically: in this a polit- 
ical or religious bias could find ready expression. In a still 
higher degree was this the case when men began to discuss 
contemporary political questions in the newspapers and 
to employ them *as a medium for disseminating party 
opinions. 

This took place first in England during the Long Parlia- 



240 THE GENESIS OF JOURNALISM, 

ment and the Revolution of 1649. The Netherlands and a 
part of the imperial free towns of Germany followed later. 
In France the change was not consummated before the era 
of the great Revolution: in most other countries it oc- 
curred in the nineteenth century. The newspaper, from 
being a mere vehicle for the publication of news, became 
an instrument for supporting and shaping public opinion 
and a weapon of party politics. 

The effect of this upon the internal organization of the 
newspaper undertaking was to introduce a third depart- 
ment, the editorships between news collecting and news 
publication. For the newspaper publisher, however, it sig- 
nified that from a mere seller of news he had become a 
dealer in public opinion as well. 

At first this meant nothing more than that the publisher 
was placed in a position to shift a portion of the risk of his 
undertaking upon a party organization, a circle of inter- 
ested persons, or a government. If the leanings of the 
paper were distasteful to the readers they ceased to buy 
the paper. Their wishes thus remained, in the final analy- 
sis, the determining factor for the contents of the news- 
papers. 

The gradually expanding circulation of the printed 
newspapers nevertheless soon led to their employment by 
the authorities for making public announcements. With 
this came, in the first quarter of last century, the extension 
of private announcements,*^ which have now attained, 
through the so-called advertising bureauXy^^ some such or- 
ganization as political news collecting possesses in the cor- 
respondence bureaux. 

"At first, it would seem, in special advice or intelligence sheets 
which in many cases come from general agelicies (Inquiry offices, 
Bureaux of Information). Comp. F. Mangold, in Basler Jhrb., 1897. 

" Annoncen-Expeditionen. 



THE GENESIS OF JOURNALISM. 241 

By admitting advertisements the newspaper fell into a 
peculiarly ambiguous position. For the subscription price 
it formerly published only news and opinions of general in- 
terest; now, through all sorts of advertisements, for which 
it receives special remuneration, it also serves private trade 
and private interests. It sells news to its readers, and it 
sells its circle of readers to any private interest capable of 
paying the price. In the same paper, often on the same 
page, where the highest interests of mankind are, or at 
least should be, represented, buyers and sellers ply their 
vocations in ignoble greed of gain. For the uninitiated 
it is often difficult to distinguish where the interests of the 
public cease and private interests begin. 

This is all the more dangerous in that in the course of 
the past century the subjects discussed on the editorial 
page of the newspapers have grown to embrace almost the 
whole range of human interests. Statecraft, provincial 
and local administration, the administration of justice, art 
in all its manifestations, technology, economic and social 
life in its manifold phases, are reflected in the daily press; 
and since the development of the feuilleton, a good pro- 
portion of literary and even of scientific products flows 
into this great stream of contemporary social and mental 
life. The book as a form of publication — ^we may have 
no doubts on the point — loses ground from year to year. 

It is impossible to enter into these matters at greater 
length. The sole aim of this cursory survey of the mod- 
ern development of journalism has been to give, from 
the view-point of historical evolution, the beginnings of 
the newspaper press their proper setting, and at the same 
time to show how the gathering of news has been con- 
ditioned at each epoch by general conditions of trade. 

The Roman newspaper is one feature in the autonomous 
administration of the wealthy and aristocratic household. 



242 THE GENESIS OF JOURNMUSM. 

A news-clerk was kept, just as was a body-surgeon or a 
librarian. In most cases he is the property, the slave, of 
the news-reader, working according to the directions of 
his master. 

In the written newspaper of the sixteenth century there 
is exhibited the same handicraft character then domi- 
nating all branches of higher economic activity. The news- 
writer, on demand and for a definite price, furnishes di- 
rectly to a circle of patrons the news that he had gathered 
— in which proceeding he doubtless suits the amount to 
the latter's needs. He is reporter, editor, and publisher in 
one. 

The modern newspaper is a capitalistic enterprise, a sort 
of news-factory in which a great number of people (cor- 
respondents, editors, typesetters, correctors, machine- 
tenders, collectors of advertisements, office-clerks, messen- 
gers, etc.) are employed on wage, under a single adminis- 
tration, at very specialized work. This paper produces 
wares for an unknown circle of readers, from whom it is, 
furthermore, frequently separated by intermediaries, such 
as delivery agencies and postal institutions. The simple 
needs of the reader or of the circle of patrons no longer de- 
termine the quality of these wares; it is now the very com- 
plicated conditions of competition in the publication mar- 
ket. In this market, however, as generally in wholesale 
markets, the consumers of the goods, the newspaper 
readers, take no direct part; the determining factors are 
the wholesale dealers and the speculators in news: the gov- 
ernments, the telegraph bureaux dependent upon their 
special correspondents, the political parties, artistic and 
scientific cliques, me|i on 'change, and last but not least, 
the advertising agencies and large individual advertisers. 

Each number of a great journal which appears to-day is 
a marvel of economic division of labour, capitalistic organ- 



THE GENESIS OF JOURNALISM. 243 

izationj and mechanical technique ; it is an instrument of in- 
tellectual and economic intercourse^ in which the potencies 
of all other instruments of commerce — the railway, the 
post, the telegraph, and the telephone — ^are united as in a 
focus. But our eyes can linger with satisfaction on no spot 
where capitalism comes into contact with intellectual life; 
and so we can take but half-hearted pleasure in this ac- 
quisition of modem civilization. It would indeed be diffi- 
cult for us to believe that the newspaper in its present de- 
velopment is destined to constitute the highest and final 
medium for the supplying of news.^* 

" [Comp. Mr. Alfred Harms worth on The Newspapers of the Twen* 
tieth Century in North Amer. Rev., Jan. 1901, — £d.] 



CHAPTER VII. 

UNION OF LABOUR AND LABOUR IN COMMON. 

There is in Germany perhaps scarcely a modem text- 
book or contemporary course of university lectures on 
political economy in which some mention is not made of 
the principle of umon of labour and some remark offered 
thereon. No one has really much to say about it. Yet it 
is there, and has its traditional place after the section on 
the division of labour, where, if it be thought at all worthy 
of it, it receives with regularity its paragraph, to come to 
light no more in the later text of the book or lecture. 

And so it has been for more than half a century. But as 
science cannot be lenient with concepts that are not fitted 
to give a deeper insight into a series of phenomena, simply 
because they have once gained currency, it is at length 
time for a closer investigation of this ancient inventorial 
item in order, if it is really unserviceable, to discard it, or to 
assign it its proper place should it be found useful in fur- 
thering our knowledge. 

According to the text-books, union of labour is nothing 
more nor less than "the other side of division of labour;" or 
" division of labour viewed from the standpoint of the or- 
ganizing unit;" ^ the " correlative of division of labour;"* 
" the reverse side of the medal whose obverse side is the 

* Both in Philippovich, Grundriss d. polit. Oek. (2d cd.), p. 78. 

* Mangoldt, Grundriss d, VolkswirthschaftsUhret § 29. 

344 



UmON OF LABOUR AND LABOUR IN COMMON. 245 

division of labour/* • These are all somewhat vague ex- 
pressionSy which on the whole seem to have their origin in 
the view that if labour is divided it must also be reunited, 
since the separate parts cannot exist independently. Here 
then the idea of division of labour is either conceived of 
very rcstrictedly — somewhat after the manner of Adam 
Smith's pin-manufactory — in which case the unifying force 
is supplied by the capital of the entrepreneur; or the con- 
ception is broadened to embrace the so-called social 
division of labour, in which case the labour-uniting element 
must be furnished by trade; so that union of labour would 
be synonymous with commercial organization generally. 

In fact, Roscher, who gives the most detailed treatment 
of the subject, and to whom all later writers resort, re- 
garded the subject in this light/ Division of labour and 
union of labour are in his view, '^ but two different aspects 
of the same conception, namely, social labour: separation of 
tasks so far as they would incommode one another, and 
their union in so far as they aid one another." " The vine- 
dresser and the flax-grower," he continues, " would neces- 
sarily die of hunger if they could not count for certain on 
the grower of grain; the workman in the pin-factory who 
merely prepares the pin-heads must be sure of his comrade 
who sharpens the points if his labour is not to be entirely 
in vain; while the labour of a merchant simply cannot be 
conceived of without that of the different producers be- 
tween whom he acts as intermediary." 

The whole phenomenon is thus shrouded in the mists of 
processes of commerce and of organization; it is made 
synonymous with economy generally. In particular it alto- 
gether loses its correlation with the notion of division of 

 Klein wachtcr, Die volksw. Produktion in Schonberg's Handbook, 

I 13- 

* Principles of Political Economy (New York, 1878), I, pp. 203-4. 



24^ UNION OF LABOUR AND LABOUR IN COMMON. 

labour. For the rest, Roscher discusses at length only the 
constancy of the progress of civilization — which is reaazed 
by each generation leaving to its successors the augmented 
inheritance of its predecessors; and further, the advantage 
of large undertakings and the association of small ones 
whereby labour ultimately disappears almost completely 
from the horizon. 

Roscher in all this goes back to Frederick List,' who, 
in his theory of the development of national productive 
powers, was the first in Germany, as far as I am aware, to 
use the expression " union of labour." Moreover, he 
turned it to peculiar account. Starting from a criticism of 
the " natural law " of division of labour, neither Adam 
Smith nor any of his successors have, in List's opinion, 
thoroughly investigated the essential nature and character 
of this law or followed it out to its most important conse- 
quences. The very expression " division of labour " was 
inadequate, he says, and necessarily produced a false con- 
ception. He then continues: "It is division of labour if 
one savage on one and the same day goes hunting or fish- 
ing, cuts down wood, repairs his wigwam, and makes ready 
arrows, nets, and clothes; but it is also division of labour 
if, as in the example cited by Adam Smith, ten different 
persons share in the different occupations connected with 
the manufacture of a pin. The former is an objective, and 
.the latter a subjective division of labour; the former hin- 
Klers, the latter furthers production. The essential differ- 
•ence between the two is that in the former one person 
♦divides his work so as to produce various objects, while in 
the latter several persons share in the production of a single 
object." 

" Both operations, on the other hand," we read further, 

'The National System of Political Economy (London, 1885), pp. 149 ff. 



UNION OF LABOUR AND LABOUR IN COMMON. 247 

" may with equal correctness be called a union of labour; 
the savage unites various tasks in his person, while in the 
case of the pin-manufacture various persons are united in 
one work of production in common. The essential charac- 
ter of the natural law from which the popular school ex- 
plains such important phenomena in social economy, is 
evidently not merely a division of labour but a division 
of different commercial operations among* several indi- 
viduals, and at the same time a federation or union of 
various energies, intelligences and powers in a common 
production. The cause of the productiveness of these opera- 
tions is not merely the division^ but essentially this union.'' 

This latter List develops further, and upon it endeavours 
to base the demand for the establishment of a harmony of 
the productive powers in the nation. The highest division of 
occupations and the highest unification of the productive 
powers in material production are found in agriculture 
and manufacturing. " A nation devoting itself exclusively 
to agriculture is like an individual engaged in material 
production with one arm gone," etc. 

Free these explanations from the ingenious rhetoric of 
the great agitator, and we find, as so often, that he has 
been unjust towards Adam Smith. The latter in no way 
overlooks the fact, as List is frank enough to admit, that 
division of labour postulates a cooperation of forces; and 
at the close of his celebrated chapter on division of labour 
he explains expressly that by means of this joint labour 
the meanest person in a civilized country may attain a 
more ample accommodation than an African king.* But 
he was keen-sighted enough not to regard this fact, which 
was involved in the nature of division of labour and iden- 
tical with it, as an independent economic phenomenon. 

*Book I, Chapter I, towards end 



348 UNION OF LABOUR AND LABOUR IN COMMON. 

What purpose would it serve to call the same thing at one 
time division of labour and at another union of labour, ac- 
cording as it was viewed from one side or the other? In a 
young science that would have been only a source of con- 
fusion. 

Of course the procedure of the Indian who successively 
hunts, fishes, fells trees, etc., would never have been recog- 
nised by Adam Smith as a particular instance of division 
of labour. On the contrary, he would have designated it 
undivided labour,'' a condition such as preceded division 
of labour throughout society. Division of labour is for 
him something else than division of time. 

Of the factor of time in the disposal of labour List speaks 
more at length in another place.® He there explains that 
the individual branches of industry in a countr>'^ only 
gradually gain possession of improved processes, machin- 
ery, buildings, advantages in production, experiences and 
skill, and all those details of information and connections 
that insure to them the profitable purchase of their raw 
material and the profitable sale of their products. It is 
easier, he believes, to perfect and extend a business already 
established than to found a new one; easier to produce 
superior goods at moderate prices in a branch of industry 
long domiciled in a country than in a newly-established 
one. " As in all human institutions, so in industry there lies 
at the root of important achievements a law of nature that 
has much in common with the natural law of the division of 
labour and of the federation of productive forces. Its es- 
sential feature consists in several successive generations 
as it were uniting their forces towards one and the same 
end, and as it were dividing among them the expenditure 
of energy necessary to its attainment." List calls this the 

* On his conception of division of labour, compare following chapter. 

* As above, pp. 294 ff- 



UNION OF LABOUR AND LABOUR IN COMMON. 249 

principle of stability and continuity of work, and seeks to 
prove its operation in history by a series of examples; the 
superiority in strength of a hereditary over an elective 
monarchy, the transmission of the acquisitions of human 
knowledge through printing, the influence of the caste 
system upon the maintenance and development of indus- 
trial skill, the building of cathedrals in the Middle Ages 
during several generations. The system of public debts by 
which '* the present generation makes a draft on a future 
generation " is also cited as a peculiarly apt instance of 
the application of the principle of continuity in work. 

It is easily seen that List here is dealing only with a 
rhetorically clad analogy to union of labour. This, how- 
ever, has not prevented later writers from forming out of 
" continuity of work " a special type of union of labour, 
although a little reflection might have taught them that it 
is a phenomenon not at all peculiar to economic activity. 
Continuity of work is the universal historical principle of 
social development by which man is distinguished from the 
animal. With each lower animal begins anew a similar 
existence which runs its course, so far as we know, to-day 
as thousands of years ago, leaving not a record, not a trace. 
But each human generation takes over the fruits of the 
civilization of all preceding generations, and hands them 
down with an increase to the succeeding age. This is true 
not merely of material production but also of art, science, 
religion, law, and custom. Continuity of work thus forms 
one of the essential conditions and first postulates of hu- 
man existence; and there is no reason for giving it special 
treatment in the theory of the economic employment of 
labour, particularly since it oflfers for the latter no new 
instructive points of view. 

Sundry text-books recognise still a third type of union 
of labour, which is said to arise " when several do 



' 



«50 UNION OF LABOUR AND LABOUR IN COMMON. 

similar work concurrently, and by virtue of union ob- 
tain a greater result than would be possible to them work- 
ing individually." Heinrich Rau, who incidentally men- 
tions this case,*^ instances temporary companies of forest 
wood-cutters, of raftsmen, and of reapers. In reality he 
here singles out a procedure that is not division of labour, 
although an increased productivity in the labour of the 
individual results from the simultaneous cooperation of 
several. This case then, like the one of the varied activity 
of the Indian mentioned by List, cannot be summarily 
dismissed as already embraced under the conception of 
division of labour and as ill-adapted to special scientific 
treatment. 

Without doubt the real reason for the formation of the 
concept union of labour and for its long retention in the 
literature of the science is the vague feeling that there 
must be an economic principle forming the counterpart of 
diznsion of labour. Cooperation it cannot be, for that is 
identical with certain forms of division of labour,*^ its 
" other side." What then is this principle? 

All diz/ision of labour is an accommodation of work to 
limited human capacity. It takes place when a qualitative 

*Gfvndsat8€ der Volkswirthschaftslehre, I, § ii6 (a). Rau appeals to 
Gioja, who had studied the matter somewhat in his Nuovo prospetto 
deUe sciense economiche, I, 87 ff. Moreover Hermann, Siaatsw, Unter- 
suchungen, new edition, p. 217, had also given it some attention. He 
designates it as "the simplest combination of labour." Similarly by 
the French who distinguish co-oph'otion simple and co-opSratian complexe, 
and make the latter identical with division of labour. Comp. Cauwes, 
Caurs d^Econ, poL, 1, § 225. 

^ For example, subdivision of labour and division of production, but 
not division of trade. If various specialists take the place of a general 
practitioner no cooperation takes place amongst them either in com- 
mercial dealings or in any other way similar to that amongst the 
different workers in a factory. 



-^-i 



UmON OF LABOUR yiND LABOUR IN COMMON. 251 

disproportion exists between the work to be done and the 
individual's capability.^ ^ 

But there may also be a quantitative disproportion be- 
tween the two factors in two ways: (i) the quantity of 
work to be done may be less than the available labour- 
power, and it may also (2) be more than equal to the 
strength of a single individual. 

In the arst case the physical force would not be com- 
pletely utilized if the labourer confined himself to this one 
line of work. His capacity for work would in part lie fal- 
low, and an uneconomic squandering of strength would 
result. The work in question could not, perhaps, form the 
basis for a life-supporting vocation. The labourer, even 
in his own private interest, will be driven to help himself 
by combining or uniting with the first a second activity to 
fill out his leisure time. We may suitably call this union 
or combination of labour}^ 

In the second case the individual by himself cannot pos- 
sibly perform the task, or can do so only with a dispropor- 
tionate expenditure of time and energy. A single work- 
man, for instance, might, if necessity demanded, succeed in 
cutting the trunk of a large tree into boards with a hand- 
saw; but with what trouble and needless expenditure of 
time! If two men and a whip-saw are called into service, 
however, the work goes forward not only absolutely but 
relatively better. The picture of the saw-pit then arises, 
which at times can still be seen in rural timber-yards. The 
union of the workers organizes the labour of each indi- 
vidual more productively. If we are to avoid the most 
lamentable confusions we must no longer designate this 
procedure union of labour,*' but at most union of labour- 

"Comp. following chapter. 

^ Arhiitsvereinigung, 

^To distinguish from the first case we would then have to say 



as* UNION OF LABOUR AND LABOUR IN COMMON, 

ers. More accurate does it seem, especially in view of the 
varieties of this process to be mentioned later, to employ 
the expression labour in common} *^ In this phrase the 
personal element, which here comes into prominence, is 
more clearly expressed. 

Union of labour is then the union of different classes of 
work in one person; labour in common is the concurrent em- 
ployment of several workers in the accomplishment of one task. 
In union of labour the same producer turns out various 
products or combines production with trading or with 
personal service; in labour in common various labourers 
produce in common the like product. In the one case the 
uniting point is in the subject of the work, the labourer, in 
the other the community lies in the object of the labour. 

The two processes are independent of each other and of 
division of labour. They, of course, play their chief 
role during primitive stages of development and in the 
lower strata of economically organized society. Two 
great stages in the economic life of nations might indeed 
be distinguished: a lower one, in which the principle of 
union of labour and labour in common comes preeminently 
into play; and a higher one, with the principle of division 
of labour predominant. In the same way two spheres of 
social existence may be distinguished in contemporary 
economic life: one with pronounced division of labour, the 
other with union of labour and labour in common. 



In a separate consideration of each of these two phe- 
nomena we had better begin with union of labour. It appears 
early in the history of peoples. It is universally met with 
directly the stage of individual search for food is passed, 

"subjective " (personal) union of labour; while the first case would be 
designated as " objective " (material) union of labour. 
" ArbeiUgemeinschaft, 



^^: 



UmON OF LABOUR AND LABOUR IN COMMON. 253 

and when economic motives, be they even of the crudest 
kind, become discernible in men's transactions. For at 
that point we everywhere notice the sharp separation of 
two distinct spheres of production, each of which again 
contains many subdepartments. One embraces men's 
work, the other women's work.^* Essentially the same 
arrangement, with unimportant variations in detail, is 
found among all more advanced primitive peoples, and we 
cannot deny that there is a certain instinctive system about 
it. Of a division of labour between man and wife one can- 
not seriously speak, for from all we know none of the occu- 
pations assigned to either of the two sexes has ever been 
carried on by the other. ' 

It must be assumed that this condition of things de- 
veloped quite naturally. In any case the statement is false 
that the stronger man " imposed " upon the woman the 
tasks falling to him. Much rather is it correct to say that 
each sex has of its own impulse — it might perhaps be said 
under the stress of environment — created in the course of 
time its own department of production, developed the 
technical details connected with it, collected the experi- 
ences, and transmitted then\ to the following generation of 
the same sex. Thus these two combinations of tasks, 
through continued hereditary transmission within the 
same sex, have almost been evolved into sexual character- 
istics or functions. The hereditary task of the woman, in 
which the man was not instructed, formed a species of 
natural equipment that made her valued by the man and 
gave her a price. Though it is true that from this 
grew the conception of the wife as property of the husband, 
it is none the less true that the important part played by 
the wife in production has been not the least important 

" A detailed discussion of these, pp. jo ff., 55. 



254 UNION OF LABOUR AND LABOUR IN COMMON. 

factor in the gfradual elevation of the rude primaeval union 
of the sexes to a community of life in which the woman 
has finally raised herself to equality of rights with the hus- 
band. 

The economic importance of the union of various tasks 
in the hands of each sex is essentially of an educational and 
disciplinary nature. It compelled as it were of itself, at 
least on the part of the wife, attention to the elements of 
time at seed-time and at harvest, and finally to a di- 
vision of time, very crude though it was, for the single day. 
It is a matter of particular moment in this connection that 
the preparation of grain by means of the primitive rubbing- 
stone, which is the method employed by most primitive 
peoples down to the present day, makes exceedingly heavy 
demands upon the operator's time, so that the mere main- 
tenance of three or four persons required the labour of one 
woman.** This is one of the most important supports of 
polygamy among these peoples, and renders it tolerable 
for the wife. For a new spouse brought home by the hus- 
band always appears to the other wives as a helper to 
lighten their lot. It is thus comprehensible that the posses- 
sion of numerous wives must serve as an indication of 
wealth. We may even assert that the careful employment 
of time, with which systematic economic action first be- 
gins, finds its starting-point in the union of labour on the 
part of the women. 

Moreover, when in the course of subsequent develop- 
ment considerable shiftings took place in the boundaries 
of the spheres of work of both sexes, forcing the wife ever 
more towards the side of supervising consumption within 
the household and placing almost all the production in 
the hands of the husband, the principle of division of 

*• Comp. Dr. W. Junker's Travels in Africa, II, pp. 170, 171, and my 
Arbeit u, Rhythmus, pp. iS, 60, 61. 



UNION OF LABOUR AND LABOUR IN COMMON. 255 

labour made itself felt almost solely in man's sphere of 
work, while to the wife's household management remained 
the most varied duties of preparing, disposing, cleaning, 
and repairing. The course of these latter really deter- 
mines the division of time in our daily life. 

To be sure, union of labour has not on that account dis- 
appeared completely from the economic world. Agrkul- 
ture still embraces occupations varying greatly from one 
another; everywhere in civilized lands its development has 
been intimately connected with cattle-raising, while sub- 
sidiary industries are often included within its sphere. 
It is indeed one of the most important tasks of the farm- 
director so to arrange matters that the working powers of 
man and beast can be turned to full account in as many 
wa3rs and in as regular a manner as possible. In the change 
of activity following the seasons of the year there is, even 
in large agricultural undertakings, but little room for sub- 
division of work and specialization; different kinds of oc- 
cupations must always be united in one person, and among 
the women workers a clear division into farm hands and 
household servants is not feasible. 

Similar considerations hold for forestry, in which keen 
practical men condemn the system still common in many 
places of having specialized labour for each season,^ ^ and 
demand the employment throughout the whole year of a 
permanent staff of all-round workmen. Such a require- 
ment can be met only on the basis of union of labour. 

In industry handicraft has from time immemorial been 
founded on union of labour. It was not the highest pro- 
ductivity that determined the mutual delimitation of the 
departments of production, but regard for the daily bread 
which each master was to find in his craft. The number- 

^^Comp. Fr. Jentsch, Die ArheiterverhaU, in d. Farshvirtk, d. Stoats 
(Berlin, 1882). 



* ... * 



25^ UNION OF LABOUR AND LABOUR IN COMMON. 

less disputes between different guilds as to the limits <A 
their trade which fill the pages of industrial history during 
the last few centuries continually raised discussions on the 
practicableness of this or that combination. In the age 
of industrial freedom handicraft has also advanced in the 
large cities in the direction of specialization; in the smaller 
towns the old combinations have been retained, and in 
country parts new ones are still arising each year. The ma- 
son is here often plasterer, painter, and paper-hanger as 
well, while m winter he serves for wage as butcher; the 
smith is at the same time locksmith and chief engineer of 
the threshing-machine; paper-hanging is cared for now by 
the saddler, now by the painter, now by the bookbinder. 
In the towns the greatest variety of combinations are 
made by the new occupations. Gas-fitting and plumbing 
are undertaken now by the locksmith, now by the tinsmith, 
and electric services are installed in houses by craftsmen of 
most diverse types. Everywhere the craftsman appears 
willing to add to his workshop a small counter • trade, 
especially with wares of his own department of labour no 
longer produced by hand-work, but often with various 
other articles as well. Justus Moser long since remarked 
the sound economic idea realized in this union; and would 
willingly have seen all petty retailing transferred to hand- 
workers and their wives.*® If we add to this the various 
alliances that handicraft makes with services of a personal 
character, especially minor civil offices, and in the country 
as a regular thing with agriculture, we are readily con- 
vinced that the union of labour still commands a very ex- 
tended field.*® Men of " modern mind " may deplore the 



"* PatrioHsche Phaniasien, Vol. II, No. XXXVII. 

"Copious material on the combinations of trades and secondary 
occupations of handworkers is oflfered in Untersuchungen iiber d, Lagi 
d, Handwerks in Deutschland, edited by myself, in Schriften d. Vcreins 



UNION OF LABOUR AND LABOUR IN COMMON. 25? 

great number of such " backward trades "; pessimists may 
see in them a sign of the '' distress in handicraft " ; fanatics 
on production may regret that under such conditions the 
highest possible measure of productivity is not realized in 
every branch of industry. But an unprejudiced judgment, 
based upon an investigation of the facts, will find that in 
the union of labour the middle class of small independent 
workmen has its firmest footing; it will find too that in 
the majority of cases the due observance of sound business 
principles has not been wanting. For as a rule it is really 
a question of making use of time that is not taken up by 
the chief occupation, and of giving employment to capabil- 
ities that would otherwise lie dormant. 

Union of labour is relatively still more common in house 
industry where the women employed at the same time at- 
tend to their domestic duties, and where the men often 
follow agriculture or some other business as primary occu- 
pation. Indeed the origin of many commission industries 
rests finally upon the consideration that persons not fully 
occupied could profitably combine them with their other 
business. 

Trading primarily is always union of labour, since in the 
earlier stages of its development it regularly includes 
transportation. Caravan trade is an example. In modem 
commercial life division of labour has strongly asserted it- 
self in wholesale trade, and also in the retail trade of large 
cities. But along with these are numerous businesses, 
such as hardware and house-furnishing shops, which bring 
together the most varied articles. In the wholesale ware- 
houses and export businesses, in the sixpenny bazaars and 
cash stores this development has reached its highest point. 
These giant undertakings of course lie beyond the range 

f. Socialpolitik, Vols. LXII— LXX, especially in the descriptions of in- 
dustries in small towns and country districts. 



258 UNION OF LABOUR AND LABOUR IN COMMON. 

of our Study, since with them the work is generally ar- 
ranged in strict accordance with the principle of division 
of labour. On the other hand, the numerous small retail 
businesses carried on in suburban places, in small towns, 
and in the country usually as the sole occupation of one 
person lie within its survey, because here the owner 
deals in every possible article that will bring in money. 
Indeed one would have to write a detailed account of the 
sale shops to explain all that is to be found gathered to- 
gether there. Certain wares are specially prized for filling 
out the stock, such as canes, cigar-holders, combs, brushes, 
and straw hats; and it is often difficult to learn how they 
have come into the company they keep. Many tradesmen 
of such a class at the same time carry on commission busi- 
nesses, insurance and news agencies, sell lottery and 
theatre tickets, receive advertisements and savings-bank 
deposits, and the like. 

In the great world of commerce there are various spe- 
cialized occupations that can hardly be carried on with 
profit by themselves, and therefore are always followed in 
conjunction with another pursuit. What village could 
support a special precentor, village clerk, or sexton; what 
rural loan association maintain a treasurer; what insurance 
company pay its army of sub-agents sufficient for their 
support? Without the possibility of union of labour these 
and many other economic functions would simply have to 
remain unperformed. 

The consideration determining the combination in each 
case could only be gleaned from a minute statistical and 
descriptive investigation. In most cases the influence 
that decides the person devoting himself to different 
kinds of work is the full employment of his time and the 
gaining of a full livelihood. For the method of combina- 
tion, however, many other considerations come into play. 



UhllON OF LABOUR AND LABOUR IN COMMON. 259 

Now it is to take advantage of a clientele already existing, 
now to utilize a particular talent or skill possessed by the 
workman for a further object. The economic principle 
will in these cases in one way or another always come into 
play. 

The actual extent of union of labour in national economy 
is not easy to determine. Statistics have sought to answer 
the need by creating the rather unsatisfactory category of 
the auxiliary occupations; but it is easy to see that this 
designation does not exhaust the total number of cases 
that come here into question; it gives at most only those 
in which the auxiliary occupation ranks in some degree as 
an independent vocation. A union of occupations might be 
spoken of in this case.^° Yet some conception of the im- 
mense economic importance of the union of labour may be 
gained when we learn from the results of the last German 
industrial census that on June 14, 1895, there were almost 
five million persons in the German Empire who had some 
secondary occupation, and that agriculture alone was an 
auxiliary pursuit for 3,648,237. Of 3,999,023 proprietors 
and managers in some branch of agriculture, industry, or 
trade, 36.9 per cent. (1,475,023) had an auxiliary occupa- 
tion, while 2,928,530 carried on these branches as auxiliary 
work. 

The following table gives a survey of the whole field of 
industrial activity covered by that census. In it those car- 
rying on independent and dependent work are grouped 
together. 

From this table we find that out of every 100 persons 
pursuing their chief occupation in one of the classes indi- 



''On the occurrence of combined occupations in town life in the 
Middle Ages there are some details in my Bevolkerung wm Frankfurt, 
I, pp. 232 ff., 417 ff. 



26o UNION OF LABOUR AND LABOUR IN COMMON. 



DIVISIONS OF THE INDUSTRIAL POPULATION OF THE GERMAN EMPIRX 
ACCORDING TO CHIEF AND AUXILIARY OCCUPATIONS ON 14TH 
OF JUNEy 1895. 





No. of Per- 
sons finding 
their Chief 
Occupation 
in Class 
indicated. 

t 


No. of Persons having 


No. of 

Persons 

making 

their 

Occupation 

indicated 

in Col. I 

Secondary. 

4 




ClSMOf 

OocniMaioii, 


a 
Secondary 
Occupa- 
tion. 

• 


no 

Secondary 

Occupation. 

3 


Toul No. 

in the 

Various 

Divisions. 

S 


A. Agriculture, 
forestry, stock- 
raising, fishing.. 

B. Mining and in- 
dustries 

C. Trade and com- 
merce .,......•. 


8.292,692 
8,281,220 
2.338.511 

43a.49i 
1,425,961 


1,049,542 

1,491.865 

384,105 

31,333 
115,266 


7,243.150 

6.789.355 
1,954,406 

401,158 
1,310,695 


3.648,237 
619.386 

569,877 
16,765 
95.436 


11,940,929 
8,908,606 
2,908,388 

449.256 

1.521,397 


D. Domestic ser- 
vice, hired labour 
of various kinds. 

E. Public service, 
liberal profes- 
sions .......a... 




Total 


20.770.875 


3,072,111 


17.698,764 


4,949,701 


25.720,576 


Of these were : 
Males 


15.506.482 
5.264,393 


2,816,655 
255.456 


12,689,827 
5,008,937 


3.203.375 
1.746,326 


18.709.857 
7,010,719 


Females 



tt 



u 



« 



<l 



cated, whether as proprietor or workman in any capacity, 
a second or third (auxiliary) occupation was added — 

In agriculture, forestry, stock-raising, fishing, by 12.6 persons 

In mining and industry by 18.0 

In trade and commerce by 16.4 

In domestic service and hired labour of various kinds by. 7.2 

In public service and liberal professions by 8.1 

Of the total number of persons following an occupation 
(cither chief or auxiliary) in one of the said classes, a 
secondary pursuit from the same class was chosen — 

In agriculture, forestry, stock-raising, fishing, by 30.6% 

In mining and industry by 6.7% 

In trade and commerce by • 19.6% 

In domestic service and hired labour of various kinds by 3-7% 

In public service and liberal professions by 6.2^ 



UNION OF LABOUR AND LABOUR IN COMMON. 261 

Even from the returns of occupations, which unfortu- 
nately are not sufficiently detailed for this purpose, it is 
clear that many occupations are carried on chiefly in con- 
junction with othfer pursuits. For example, in the table 
next presented the total number of persons following the 
vocation are contrasted, by percentages, with those fol- 
lowing it either as principal or subsidiary, along with an- 
other occupation.*^ 



Stock-raising 83.47<^ 

Inland fishing , 69.3% 

Turf cutting and preparing 93-9% 

Stone cutting and carving 57-2% 

Marble, stone, and slate quarrying and the manufacture of rough 

wares from these materials 78.6% 

Manufacture of fine stone wares 50.2% 

Brick and clay-pipe making 86.9% 

Butchering 58.1% 

Insurance 68.7% 

Personal transport and post 53-2% 

Freight delivery 75-7% 

Pottery 57-5% 

Manufacture of earthenware and glass toys 56.0% 

Nail-manufacture 67.0% 

Blacksmithing 76.8% 

Wagon-making 74-8% 

Flaying 85.9% 

Charcoal-burning 81.2% 

Flour-milling 91.6% 

Baking 61.6% 

Turning 52-7% 

Advertisement and labour-agency work 54-4% 

Inn-keeping and refreshments 64.4% 

These figures naturally are far from giving a true picture 
of the results of combined and divided labour in the oc- 
cupation classes indicated. It is at once apparent that in a 
statistical return on extractive production a country shoe- 
maker who devotes a quarter of his time to land-cultivation 

** Columns 6 and 8 of Reichsstatistik, Tab. I. 



263 UmON OF LABOUR AND LABOUR IN COMMON. 

has necessarily the other three-quarters of his time left 
out of account. This, however, is not the point; it is 
rather the question of the number of human beings to 
whom a combination of occupations assures more abun- 
dant sustenance and also in most cases a more satisfying 
existence, both as regards health and morals, than a one- 
sided employment in full agreement with the principle of 
division of labour. In the German Empire this num- 
ber is large beyond expectation, amounting almost to one- 
third of all persons engaged in earning their living. 

The principle of union of labour, despite the wealth of 
forms in which it appears, is quite simple: how shall surplus 
strength be productively employed? The principle of 
labour in commoti cannot be reduced to such a smooth for- 
mula. In general, its aim is so to supplement the insuffi- 
cient strength of the individual that the task presented can 
be accomplished. But the individual workman's insuffi- 
ciency of strength may again have different causes. It 
may be based on a definite mental disposition that pre- 
vents the workman labouring continuously by himself; it 
may rest on lack of bodily strength; or, finally, it may re- 
sult from technical conditions rendering it impossible 
for one piece of work to be performed unless accompanied 
by a second of a different character. Each of these three 
cases produces a distinct kind of labour in common. The 
first may be called companionship or fraternal labour, the 
second, labour aggregation, and the third, joint labour. We 
will consider them in order. 

I. Companionship or fraternal labour ^'^ occurs when sev- 
eral workers come together and labour without the indi- 
vidual becoming in the progress of his task in any way 

^ GeseUschaftsarbeit or gesellige Arbeit. 



UNION OF Lj4BmUR AND LABOUR IN COMMON. 263 

dependent upon the others. Each thus labours for himself 
independently and adopts any tempo he pleases. The sole 
aim in union is to have the company of fellow-workmen, to 
be able to talk, joke and sing with them, and to avoid soli- 
tary work alone with one's thoughts. 

The student whose work thrives best in undisturbed soli- 
tude will on hearing this probably shrug his sympathetic 
shoulders in pitying contempt, and find the subject hardly 
worth serious consideration. But anyone who has ever 
observed a group of village women braking flax, or doing 
their washing at the brookside, or watched a troop of 
Saxon field- workers hoeing turnips, or a line of Reapers at 
work, or listened to the singing of a group of house-paint- 
ers, or of women at work in an Italian vineyard, will be of 
a different opinion. The lower the stage of a man's cul- 
ture the more difficult is it for him to stick to continuous 
and regular labour, if he is to be left by himself. 

But the best proof of the importance of fraternal labour 
lies in fts having found some sort of organization in all 
parts of the earth. We may call to mind the public work- 
ing-places and common-houses of the savages,^* the com- 
mon workrooms of the house-workers in Russia, the spin- 
ning-rooms of our peasant girls which the bureaucracy of 
the eighteenth century so fatuously opposed, but which 
live on in many villages to the present day in the evening 
gatherings for work in common. Custom everywhere 

"K. V. d. Steinen, Unter d. Naturvolk, BrasiL, p. 374. Erman ift 
Ztschr. f. Ethnol., II, p. 378 (on the Coljusches in Sitka). Jacobsen,. 
Reise in d. InselweU d, Banda-Meeres, p. 213. Finsch, Samoafahrten, p. 357. 
Burton, as above, pp. 54, 297, 461. Nachtigal, Sahara u. Sudan, II, 
p. 624; III. pp. 146, 244. Count Schweinitz, Durch Ostafrika im Krieg 
u. Frieden, p. 171. Stanley, Through the Dark Continent, II, p. 82, and 
How I found Livingstone, pp. 546 ff. Semon, In the Australian Bush, 
p. 324. Corap. also Arbeit m. Rhythmus, pp. 38, 39, 71; and above, 

p. 37' 



a64 UNION OF LABOUR AND LABOUR IN COMMON. 

joins to these gatherings dancing, feasting or other 
practice tending to inake the work more agreeable. A 
few instances will serve to show the wide extent of such 
institutions. 

In the Fiji Islands " several women always unite in the 
preparation of tapa; frequently all the women of the place 
sit together," and in net-fishing " women always work to- 
gether in small groups; the work is at the same time a 
recreation, and there is often a merry time in the cooling 
bath.*' ^* Among many negro tribes in Africa women can 
be seen at public work-places pounding or grinding corn 
in common. A more circumstantial report is given by a 
missionary as to the North American Indians: ^' " The till- 
ing of the ground, getting of the fire-wood, and pounding 
of com in mortars is frequently done by female parties, 
much in the manner of . . . husking, quilting, and other 
frolics. . . . The labour is thus quickly and easily per- 
formed; when it is over, and sometimes in intervals, they 
sit down together to enjoy themselves by feasting on some 
good victuals prepared for them by the person or familj 
for whom they work, and which the man has taken care 
to provide beforehand from the woods; for this is consid- 
ered a principal part of the business, as there are generally 
more or less of the females assembled who have not, per^ 
haps, for a long time tasted a morsel of meat, being either 
ividows or orphans, or otherwise in straitened circum- 
.stances. Even the chat which passes during their joint 
labour is highly diverting to them, and so they seek to be 
•employed in this way as long as they can by going round 
to all the villagers who have ground to till." 

The same linking of work and pleasure is found in the 

^A. Bassler, Sudsee-Bilder, pp. 224-6. 

* Heckewelder, as above, p. 156- Similar report from South Amer- 
ica in Ehrenreich, Beitrage s, Volkerkund, BrasiL, p. 28. 



UNION OF LABOUR AND LABOUR IN COMMON. 265 

public social houses that are met with almost universally 
among primitive peoples. Divided regularly according to 
the sexes, they are most frequently built for unmarried 
men and girls. They serve not merely as places of resort 
for common tasks, but often as sleeping places as well, and 
always as places for dancing and play. There singing and 
joking and chatting go on ; the fruitless efforts of the awk- 
ward are ridiculed and the successful work of the diligent 
and skilful applauded. 

A distant parallel to this institution has been retained 
among ourselves almost down to the present in the spin- 
ning-rooms of peasant girls.^* These rooms had in every 
part of Germany their well-defined though unwritten rules 

** But with similar conditions these are met with everywhere. Henry 
Savage Landor (In the Forbidden Land, an Account of a Journey in Tibet, 
If pp. 109 ff.) found them even in the southern Himalayas among 
the Shokas, where girls and young men come together at night in 
particular meeting-places (Ram bangs), for the sake of better acquaint- 
ance, prior to entering into matrimony. ** Each village possesses one 
or more institutions of this kind, and they are indiscriminately patron- 
ized by all well-to-do people, who recognise the institution as a sound 
basis on which marriage can be arranged. The Rambang houses are 
either in the village itself, or half-way between one village and the 
next, the young women of one village thus entering into amicable 
relations with the young men of the other and vice versa, I visited 
many of these in company with Shokas, and found them very inter- 
esting. Round a big fire in the centre of the room men and women 
sat in couples, spinning wool and chatting merrily, for everything ap- 
peared decorous and cheerful. With the small hours of the morning, 
they seemed to become more sentimental, and began singing songs 
without instrumental accompaniment, the rise and fall of the voices 
sounding weird and haunting to a degree. . . . Smoking was general, 
each couple sharing the same pipe. . . . Signs of sleepiness became 
evident as morning came, and soon they all retired in couples, and 
went to sleep in their clothes on a soft layer of straw and grass. . . . 
At these gatherings every Shoka girl regularly meets with young men; 
and while she entertains the idea of selecting among them a suitable 
partner for life, she also does a considerable quantity of work with her 
spinning-wheel." 



266 UNION OF LABOUR AND LABOUR IN COMMON. 

and laws. '* In Brunswick the spinning-rooms began with 
the approach of winter, when the field-work was ended. In 
many villages this occurred at Martinmas. They lasted 
then till Lent, or at latest till Palm Sunday, when other 
work had to be done. The evening gatherings were held 
in rotation at the houses of the different members of the 
particular weaving-circle. The membership of such a circle 
was made up of four, or at most of eight girls, who were 
friends or relatives of each other. The majority consisted 
of servants, though the daughters of peasants also joined 
in. The old folks spun by themselves. In the early part 
of the evening the girls were alone; for not till later on, 
about eight o'clock, did the male visitors, who by that time 
had finished their work, put in an appearance and take 
part in the company, at first with reserve, and then more 
and more boldly. The institution has as its basis a 
laudable diligence on the part of the girls." *^ Generally 
there was a fixed amount of work for the week reckoned in 
yam; anyone not doing it received a nickname. At times 
a spinning contest was arranged, while a feeling of lively 
emulation always prevailed.*® Indeed a species of labour 
police was maintained over the individual members. In 
the district of Nassau a moustache was painted with a piece 
of charcoal on the spinner who fell asleep; if she let the 
thread break and slip from her a lad might take her distaff, 
which she had to redeem with a kiss.** 

"R. Andree, Braunschweig. Volkskunde, pp. i68ff. Comp. K. Frei- 
herr v. Leoprechting, Aus d. Lechrain, pp. 201-2; Ztschr. d. Ver. f. 
Volkskunde, III, pp. 291, 292, VIII, p. 366; and in detail Bockel, Volks- 
tieder aus Oberhessen, pp. cxxiii ff. 

"Interesting notes on competitive games in spinning in Ztschr d. 
Ver. f. Volksk., VIII, pp. 215, 216. Comp. Arbeit u. Rhythmus, pp. 91 ff- 

"Among the Wends in Lusatia on the last spinning night before 
Christmas the slow and lazy ones are brought to trial: Haupt and 
Schmaler, Volkslieder d. Wenden in d. Ober- u. Nieder-Lausits, II, p. 22a 
Similar instances of labour supervision by comrades are found in 



UNION OF LABOUR AND LABOUR IN COMMON. 267 

The spinning-room has fallen a sacrifice to the technical 
revolutions of modem times; but all through the country 
during long winter evenings the young girls still congre- 
gate with their work in the house of a friend. This is also 
the case with several house industries carried on in the coun- 
try, for instance, with lace-making in the Erz mountains, 
where this kind of gathering of working girls is still re- 
ferred to as " go spinning." *® This practice is fully devel- 
oped in the system of house industry in Russia.** Here 
male and female Kustaris frequently work together outside 
their homes. Large companies, often composed wholly of 
house-workers of the one village engaged in a similar 
trade, gather together in a particular workroom, which is 
either a large room rented in a peasant's house or a shop 
erected for the purpose. Such a common workroom is still 
most frequently termed "spinning-room" (^w^Z/^/Aja), and 
often " factory." It is to be found, for instance, in do- 
mestic cotton-weaving, cloth-making, silk-spooling, and 
the making of shoes and toys. In women's work only 
young girls, as a rule, attend, while the married women 
work at home. 

" According to the statements of the oldest people, cot- 
ton-weaving was at first carried on exclusively in the swet- 
jolka, because the technical handling of the loom could be 
learned more quickly and easily under the constant super- 
vision of one skilled in weaving. The living-room of the 
house served at first as swetjolka, but later on a swetjolka 
separate from the house was built. To-day the young peo- 
ple and the diligent weavers still prefer to work in the 

other classes of peasant working groups. Comp, Hormann, D. Tiroler 
Bauemjahr (Innsbruck, 1899), pp. 50, 52, 66, 70, 71, 75, 129. 

" Arbeit u, Rhythmus, pp. 99, 100. 

" Details in Stellmacher, Ein Beitrag js. Darstellung d. Hausindustrie 
in Russland, pp. 106 ff. M. Gorbunoff, Uber russischc S f^ifccnindustrie 
(Vienna, 1886), pp. 23 ff. 



26S UNION OF LABOUR AND LABOUR IN COMMON. 

swetjolka rather than at home; the former because it is 
more sociable, the others because they can work more 
regularly and to greater advantage. At home the weaver 
is often called away to domestic affairs; the living- 
room there is not so spacious and bright, the air not so 
pure, since calves and lambs are not infrequently co- 
dwellers with the folks; in the swetjolka, also, the cotton, 
which at home is very liable to become moist and mouldy, 
can be better preserved." 

Thus fraternal labour accords very well with the eco- 
nomic principle, even though it originates primarily in the 
social instinct. In the company of others people work with 
greater persistence than they would alone, and in general 
because of the rivalry, also better. Work becomes pleas- 
ure, and the final result is an advance in production. 

2. By labour aggregation^^ we mean the engaging of 
several workmen of similar capacity in the performance of 
a united task, such as loading a heavy burden, shifting a 
beam, mowing a meadow, beating for the hunt. In order 
to make the employment of a plurality of workers profit- 
able, the work to be done need not be of itself too heavy 
for the strength of one person ; it is only necessary that it 
cannot be done by such an one in a reasonable time. La- 
bour aggregation is of special importance for seasonal 
work or for work that is dependent upon the weather. 
Social conditions can also play a part in expediting certain 
tasks. 

These circumstances have early led to a species of social 
organization of aggregated labour, founded on the duty 
recognised the world over of mutual assistance among 
neighbours. We may use the expression current among 
the south Slavs and call it bidden labour?^ Whenever any- 

" ArheUshduHmg, 
** Bittarbeit. 



UNION OF LABOUR AND LABOUR IN COMMON. 269 

one has work to be done for which his own household is 
not adequate^ the assistance of the neighbours is sought. 
They give it at the time without further reward than their 
entertainment, which the head of the house offers in the 
accustomed way, solely in the expectation that when need 
arises they, too, will be aided by their neighbours. The 
work is carried out in sprightly competition amid jokes 
and song, and at night there is often added a dance or like 
merry-making.'* 

This is a world-wide practice. Traces of it can even be 
found among the South Sea Islanders. In New Pomer- 
ania, for example, it is the custom for several families to 
cooperate in plaiting the fishing-baskets and large fishing- 
nets. " Before the basket receives its first dip in the water 
a meal in common is given, in which all who were engaged 
in the making participate.*' ** 

In Djailolo (Halamahera) when a piece of land belong- 
ing to a local community is to be cleared, ten to twenty 
relatives are called together to assist in felling the trees, 
their services being compensated later in other work. So 
it is at the planting of paddy, and at the rice harvest.'* 
" Whenever anyone wants to build a house he solicits some 
of his relatives to help cut the building material while the 
tide is out, he providing them the while with food. For 
the roofing, which is done with sago leaves, more helpers 
are invited. These then hold a feast, at which the chiefs 
are generally present." '• 

" Numerous instances of this custom in Chapter V of my Arhtii m. 
Rhythmns. 

•* Parkinson, Im Bisfnarck-Archipel, p. 115. 

"Riedcl, in Ztschr. f. Ethnol., XVII (1885), PP. Toff. Similarly in 
New Guinea, Finsch, Samaafahrten, pp. 56 ff. Among the Bagobos in 
South Mindanao: Schadenberg in Ztschr. f. Ethnol., XVII, pp. 19 ff. 

"Riedel, as above, p. 61. Kubary, Ethnogr, Beitrdge g. Kenntnis d, 
KaroUnen-Archipels, p. 264. C. Hose, The Natives of Borneo, in JouriL 
of Anthrop. Inst., XXIII (1894), PP. 161, 162. 



ayo UNION OF LABOUR AND LABOUR IN COMMON. 

Among the Madis or Morus in Central Africa each culti- 
vates his own land; if it is of considerable extent^ and re- 
quires more hands than his family can furnish, he calls his 
friends and neighbours to his assistance. On such occa- 
sions payment is neither given nor expected, but all are 
ready to render such help and to receive it.*^ This cus- 
tom appears to be prevalent throughout Africa; the pos- 
sessor of the land has, as a rule, to supply generous enter- 
tainment to the whole company.'® Among the Gallas the 
inhabitants of a village assemble on the threshing-place to 
thresh the panicles of the durra and to root up the 
corn amid the singing of melodious songs adapted to 
the strokes in threshing.'® Bidden labour is also common 
in building a house.*^ Among the Hovas of Madagascar 
when the grave of an important man is to be built, not the 
relatives and members of his tribe alone assist in trans- 
porting the heavy rocks, but all the inhabitants of the vil- 
lage in which he lived. There is no money-payment for 
such services; in its stead, however, great masses of pro- 
visions must be supplied during the transport of the heavy 
stones, which usually lasts many days, often for long inter- 
mediate periods; and above all many oxen must be killed. 
As the people are accustomed to help one another, no in- 
considerable part of their time is spent in such services. 
On the highways of the country one often meets great 
processions of two to three hundred men, women and 
slaves, who pull by starts on the strong ropes by which 



" Robert Felkin in Proceedings of Roy. Society of Edin. 1883/4, 
p. 310. 

"• Endemann in Ztschr. f. Ethnol., VI, 1874, P- 27. Pogge in Wiss- 
mann, Unter deutsch, Flagge quer dureh Afrika, p. 311. Nachtigal, Sahara 
u, Sudan. Ill, p. 240. Post, Afr. Jurisprudens, II, p. 172. 

" Paulitschke, Etknogr. Nordost-Afrikas, I, pp. 134, 217. 

*• Schurtz, Afr. Gewerbe, p. 21. 



UNION OF LABOUR AND LABOUR IN COMMON. 271 

means the stone, placed on a rough boat, is drawn for- 
ward.*^ 

The Georgians (Central Transcaucasia) resort to bidden 
labour at the vintage, in sowing and harvesting maize and 
wheat, in hewing and drawing wood from the forest. In 
Servia it is customary at grass-mowing, maize-cutting, 
plum harvest, the vintage, and also with spinning, weaving 
and carpet-making; in many parts of {Itissia at the hay and 
grain harvests, in hoeing turnips, felling timber, transport- 
ing manure and in ploughing, as well as with the women 
in spinning and even in scrubbing the house. In Germany 
it remains quite general in the country in house-building 
and locally in minor agricultural tasks (flax-pulling, bean- 
cutting, sheep-washing). It is an expedient of independ- 
ent household economy, as is readily seen, and recedes 
more and more with the appearance and advance of the 
entrepreneur system. 

But in most cases where bidden labour was formerly 
usual the large landowner will still engage a number 
of labourers if he is not able to hurry along the work fast 
enough with his machines. Labour aggregation becomes 
particularly important for him in the early stages of a 
process of production whose final stages can be more 
cheaply completed when carried on concurrently. A 
meadow could perhaps be mowed by one labourer in three 
days. Yet where possible the owner will employ six or 
more mowers who dispose of the work in a forenoon, be- 
cause all the grass should be dried uniformly and all the 
hay drawn in at once. Frequent drawings would increase 
the costs of production. 

Even where there are no such reasons the farmer 
whose fields are intermixed with those of others will always 

• Sibree, Madagaskar, pp. 255, 256. 



27 « UNION OF LABOUR AND LABOUR IN COMMON. 

prefer to undertake the fields in succession with all his help 
rather than divide it among the different fields. The work 
goes forward better and more briskly in company than in 
solitude, for no one will lag behind the others; moreover, 
of itself the rapid progress of the work enlivens the work- 
ers, while a piece of work in which no progress can be 
recognised, and the end is not in sight, always disheartens. 
Thus the six mowers in the instance just given will, with 
average diligence, finish the meadow not in one-sixth of 
the time that the single mower with the same diligence 
would require, but in a shorter time. Moreover with large 
undertakings in which the proprietor does not work along 
with the others, it has further to be considered that with 
dispersion of the workers the costs of supervision per unit 
of acre increase.^* 

Labour aggregation belongs almost exclusively to the 
class of work requiring little skill, and which can be exe- 
cuted with simple manual implements, or even without 
any tools whatever. It is thus found in its widest extension 

^ Even in the Grundsaize d, rationelUn Landwirthschaft (4th edition, 
Berlin, 1847), I, pp. 112 ff., of A. Thaer we find the following rules: 
" Large tasks are never to be undertaken many at a time, especially in 
places at a considerable distance. As far as possible they must be 
finished in succession, and in every case with all possible energy, 
partly on account of the superintendence, partly because a certain 
rivalry can be awakened among the workers if many of them are un- 
der supervision together. On the other hand, if the task is extensive 
and only a few of them are set to work, they become almost fright- 
ened at its extent and its slow progress and even lose courage, and 
believe that, because of its magnitude, their absence will not be no- 
ticed. In such extensive tasks one man or one team too many is 
always better than one too few. In smaller tasks, on the contrary, 
care must be taken not to employ more than are necessary. The men 
easily get in each other's way, depend upon one another, readily be- 
lieve that their work has been thought heavier than it really is." 
Similarly, H. Settegast, Dii Landwirthschaft u, ihr Betrieb, I, p. 313; 
III, p. ij8. 



UNION OF LABOUR AND LABOUR IN COMMON. 273 

in epochs of undeveloped technique,** declining as the im- 
plements of labour are improved. Yet a considerable 
sphere still remains to it. The grandest instance of labour 
aggregation is at all times presented by the standing 
armies. 

When a large number of persons labour together two 
kinds of labour aggregation are possible. In the first the 
individual workmen remain independent of one another 
during the moments their strength is called into play, and 
work together only for the more speedy disposition of the 
task. We will designate this first species simple aggrega- 
tion of labour. Instances are presented in several masons 
working on a new structure, a number of pavers on a road, 
a group of diggers or snow-shovellers, a row of mowers 
or turnip-hoers. An intermediate form is offered by a 
band of African carriers marching one after the other in 
single file, by beaters at a hunt, by several ploughers in a 
field. 

In the second case the activities of the different work- 
men do not proceed independently of one another, but 

« 

either simultaneously or with regular alternations, that is, 
they always proceed rhythmically. We will name this kind 
of labour agglomeration concatenation of labour, because it, 
so to speak, links each one taking part in the work to his 
neighbour through the succession of his movements, and 
combining all by means of the tempo into unity of system, 
makes it, as it were, an automatically working organism. 
All tasks falling under this head must, if continued for some 
time, adopt a rhythmical course. There are some, to be 
sure, that are completed with a single exertion of 

^This was the case especially among the ancient Egyptians. Some 
details on this have been gathered together in Arbeit u. Rhythmus, 
pp. 109. no. 



274 UNION OF LABOUR AND LABOUR IN COMMON. 

Strength, such as several lifting a heavy weight by word of 
command or pulling down the trunk of a tree with a rope. 

The tasks of this class, which are performed rhythmic- 
ally, can be sub-divided according as the powers of the in- 
dividual workers are exerted simultaneously or alternately 
into labours with concurrent tempo and labours with alternate 
tempo.^* 

Labours with concurrent tempo are performed, for in- 
stance, by the two lines of rowers in propelling a boat by 
oars, by sailors in heaving an anchor, in hoisting sail, in 
towing a boat against the stream, by carpenters, who, in 
laying a foundation with a pile-driver, drive great posts 
into the earth, by those drawing up barrels, and generally 
by all groups of workmen who have to move a weight by 
pulling together on a rope, by the two, four, six, or eight 
people carrying a hand-barrow or a sedan-chair, and by 
soldiers on the march. Very frequently the keeping of 
time during the work is assisted by simple counting, by a 
chorus among the workers, or by the sound of a musical 
instrument, especially of the drum. 

Examples of workmen labouring w-ith alternate tempo 
are: three stone-setters hammering in time the pavement 
stones with their paving-beetles; three or four threshers 
on the barn floor, two smiths hammering, two woodmen in 
the saw-pit or chopping a tree, two maids blueing linen or 
beating carpets. 

In tasks to be performed with concurrent tempo the 
problem is to accomplish by combination a task far surpass- 
ing the strength of one person, with the smallest number of 
labourers possible, so that all taking part in the work 



^ More in detail in my paper already frequently mentioned, Arbeit n. 
Rhythmus, to which reference may once for all be made for the follow- 
ing paragraphs as well. 



UNION OF LABOUR AND LABOUR IN COMMON. 2 75 

shall be led to apply the utmost amount of energy at the 
same moment. 

In tasks with alternate tempo we meet as a rule with 
labours that in themselves could be performed by a single 
individual. Generally they, are fatiguing tasks in which 
the various motions, such as raising and lowering the arms 
in striking with the threshing-flail, require more or less 
time. The individual worker here is always tempted to 
allow himself a brief pause for rest after each stroke or 
thrust, and thus loses the rhythm of the movements. The 
strokes or blows then succeed one another with unequal 
force and at irregular intervals, whereby the work is much 
more tiring in its results. If now a second or third work- 
man be added, the motions of each individual will regulate 
themselves by the rhythmic sound that the instruments 
give forth in striking the material worked upon. A quicker 
tempo is realized, which can be maintained with little diffi- 
culty. Each workman remains indeed independent, but 
he must adapt his movements to those of his comrades. 
The import of the matter is thus not that the magnitude 
of the task demands a doubling or tripling of forces, but 
that a single person working alone cannot maintain a 
definite rhythmical motion. 

To be sure, the sole consequence of calling in a second 
or third workman one would imagine to be the doubling 
or tripling of the effect of one workman's expenditure, yet 
this kind of labour concatenation results in a heightened 
production, inasmuch as it regulates equably for each the 
expenditure of force and the pauses for rest. The single 
workman lets his hands fall when he grows tired, or at 
least lengthens the tempo of his movements. Quick tempo 
in work enlivens ; the men working in common are stimu- 
lated to rivalry; none will fall behind the other in strength 
and endurance. 



276 UNION OF U4BOUR AND LABOUR IN COMMON. 

This pressure upon the weaker workman to equal the 
stronger becomes prominent in some tasks with a some- 
what free rhythm in which the concatenation is realized by 
grouping the workmen in rows and making the progress 
of the work of the one row dependent upon the activity of 
the other. In a line of mowers in a meadow each man 
must perform his task uniformly with the rest if he is not 
to retard the man following him or run the risk of being 
struck with his scythe. In a line of labourers handing or 
tossing each other the bricks for building, each one in a 
series must receive with equal speed if he does not want to 
bring the whole work to a standstill. 

This mutual accommodation of workmen to each other, 
which is peculiar to all kinds of labour concatenation, thus 
becomes a disciplinary element of the greatest importance, 
especially for unskilled work, such as preponderated at 
primitive stages of economic and technical development. 
It can, indeed, be instituted also as a means of discipline 
to accelerate the work in those cases of labour aggregation 
that in themselves do not require such a linking of move- 
ments. For these there are artificial means of marking the 
tempo (counting, singing, accompaniment of music), by 
means of which simple labour aggregation is changed into 
labour concatenation. This is the case with slave labour, 
which, for obvious reasons, must always be carried on by 
gangs, and with the public labours of primitive people. 

In Camerun "the chief Ngilla, a well-known Moham- 
medan slave-hunter, drew up his people in companies of 
one hundred, and had his hoeing done to the beat of music 
which followed. Behind these workmen marched the sow- 
ers to the same beats, throwing seeds from a sack which 
was hung about them," *^ The Basutos assemble every 

** Meinecke, D. deutsch. Kolonieu in Wort u, Bild, p. 35, with illustra- 
tion. 



UNION OF LABOUR AND LABOUR IN COMMON. 277 

year to cultivate the fields necessary for the personal sus- 
tenance of their chief and his principal wife. " It is a re- 
markable picture," writes the missionary Casalis,*® " to see 
on such an occasion hundreds of black men drawn up in 
straight lines moving their hoes up and down in unison. 
The air resounds with the songs by which the workers are 
enlivened, and by which they can keep the proper time. 
The chief makes a point to be present and sees to it that 
several fat oxen are killed and made ready for the labour- 
ers. All classes adopt the same plan to lighten and ex- 
pedite their work; but among the common people it rests 
on reciprocity/' 

The last example shows very clearly the transition from 
bidden to manorial labour. We find the same thing in 
the Soudan, where the erection and repairing of the village 
walls in particular is carried out to the accompaniment of 
music; and again among the Malays and the Chinese, who, 
since early times, have directed the public manorial services 
by the beat of the drum. In Europe also this means has 
been essayed. In the Baltic provinces down to the end of 
the eighteenth century landowners had their harvesting 
done by the serfs to the rhythm of the bagpipes, and 
traces of similar usage are at hand from other countries. 
In our modern States we meet with this species of labour 
concatenation brought about by artificial means only in the 
measured cadence of military forces, where the aim is al- 
ways to train a number of men to complete unanimity in 
their exercise pf strength, and where the breaking of the 
tempo by a single person detracts from the general effect. 

**Les Bassautos, p. 171, with illustration. Another can be seen in 
Gcrland, Atlas d. Ethnographic (Leipzig, 1876), Tab. XXII, No. 25. 
Similar reports by K. Endemann of the Sotho negroes in Ztschr. f. 
Ethnol., VI, pp. 26 and 30; Paulitschke, as above, I, p. 216; from the 
Gallas by Harar; and on the Bagabos in Southern Mindanao by 
Schadenberg in the same publication, XVII, pp. 19, 20. 



278 UNION OF LABOUR AND LABOUR IN COMMON. 

3. We come now to the last kind of labour in com- 
mon, which we have designated joint labour ^"^ Certain 
tasks in production require for their performance the 
simultaneous cooperation of various classes of labour. 
These latter supplement one another, and may be called 
complementary phases of labour. Since they cannot 
possibly be performed by one workmaii, several workmen 
of various kinds must be combined in one group to form 
an organized and indivisible whole. Such a group is some- 
times called gang, company, band. (In Bavaria and Aus- 
tria Pass; *® in other parts, RottCy Truppe, Bande.) 

Instances from agriculture are quite numerous. Thus in 
drawing in hay or corn, the load-builder, the pitcher, the 
after-raker, in binding, the binder and the gatherer, form 
natural groups; in mowing grain a second person is re- 
quired to glean, in digging potatoes another gathers them 
up. From the sphere of industry may be mentioned the 
smith and the bellows-man, the rope-maker and the man 
who turns the wheel, mason and hod-carrier, those placing 
and those pounding in the paving-stones; from other 
spheres: the cook and the turnspit; inn-keeper, waiter, and 
house-boy; on the street-car, driver and conductor; in the 
row-boat, oarsman and steersman, likewise hunters and 
beaters, musician and dancer, blower and organist, drum- 
mers and pipers, judge, bailiffs and clerk, doctor and at- 
tendants, a theatrical troupe, an orchestra. The list could 
be continued much further. 

In all these cases it is not a question of processes that 
have arisen through division of labour and then been re- 
united, but of activities of quite different kinds, none of 
which could ever exist by itself, and which, therefore, have 

•^ Arbeitsverbindung. 

*• Comp. Schmeller, Bayer. Worterbuch, I, p. 409. The origin of the 
words is not clear. 



UNION OF LABOUR AND LABOUR IN COMMON. 2 79 

come conjointly into being. In their advancement these 
occupations are dependent on one another, support one 
another and only together from a coordinated whole. The 
workers engaged must therefore accommodate themselves 
to each other; the one must work into the hands of the 
other, without whom he could accomplish nothing at all. 
In most cases his labour by itself would be quite unpro- 
ductive. 

As a rule there will be in such associations of labour an 
activity that can be designated the leading or dominant 
one, while the other is subordinate and auxiliary. Ac- 
cordingly the personal relationship between the workmen 
employed will also take the form of a dependent relation- 
ship. If the directing workman is independent, the work- 
man who in technical matters is dependent will frequently 
stand in the relation of employee. If the associated 
labour is made part of an undertaking it is usual for the 
whole work (" team-work," collective piece-work) to have 
assigned to it a collective wage, as is the case with the 
cigar bunch-breaker and roller, the glass-blower and at- 
tendant. The plan thus offers a means of applying the 
system of piece-wage even in cases where the work of one 
labourer cannot be separated from that of another or of 
several others; but it results in most cases to the disad- 
vantage of those who perform the subordinate labour.^^ 

On the whole, this form of labour in common belongs 
also to the stage of undeveloped technique in the instru- 
ments of production. With advancing development the 
supplanting of the subordinate labour by animal or me- 
chanical power is aimed at. The most familar example is 
offered by the plough, which was formerly drawn by hu- 
man beings, later by oxen. In this, however, the combina- 

^ Comp. Schloss, Methods of Industrial Remuneration^ pp. 6i ff. 



2So UNION OF LABOUR AND LABOUR IN COMMON. 

tion of labour endured some time longer, inasmuch as a 
second driver or several drivers were required besides the 
ploughman, until at length a more perfect construction of 
the plough made them superfluous.^® 

In conclusion, it is again to be emphasized that the whole 
sphere of labour in common belongs, like that of union of 
labour, preeminently to the departments and the epochs of 
labour possessing little or no capital. They are the re- 
source of the economically frail. As such, however, their 
great evolutionary and historical importance lies in their 
training of man to methodical division and economy of 
time, to self-subordination to a general aim, and to regular 
and intensive labour. These supplement each other in that 
the inherent weakness of union of labour, pervading 
the life of each man in primitive times, everywhere finds its 
counterpart in the temporary communities of labour that 
arise wherever the variously employed skill of the indi- 
vidual is inadequate to a gfiven task. Resting originally 
on custom alone, they lead in course of time to relation- 
ships capable of legal compulsion, such as slavery and serf- 
dom. 

The principles of union of labour and labour in common 
have in other respects contributed little to the creation of 
permanent organizations, but they have left permanent 
works. The pyramids and stone monuments of Egypt, the 

^ Interesting modifications of the system of combination of labour 
are found in the cases where more expensive implements are neces- 
sary, and only one of the parties possesses them, while the others 
merely contribute their labour. In North Russia this is particularly 
the case in fishing, and again in the work of ploughing, where the 
hitching together of six to eight animals is rendered advisable from 
the heaviness of the soil. Examples from Wales, Ireland, and Scot- 
land in Seebohm, Village Communities (4th ed.), p. 81. Meitzen, Siede- 
lung u. Agrarwesen d. Westgermanen u. Ostgermanen, d. Kelten, etc., I, 
pp. 212 ff,; II, pp. 129. 130. Similarly on the Bogos in the mountains of 
Abyssinia, in Post, Afr. Jurisprudenz, II, pp. 184, 185. 



UNION OF LABOUR AND LABOUR IN COMMON. 281 

r 

ruins of the giant cities of Mesopotamia, the structures 
of the peoples of early American civilization must be ob- 
served if we would know what human beings are capable 
of performing, even without the knowledge of iron, with- 
out draught animals, and without such simple mechanical 
expedients as lever, screw, or pulley, when united by one 
mighty mind in community of work. 

For science also the two phenomena here referred to, 
now that they have been defined, may prove themselves 
upon unbiassed testing not altogether useless building 
stones. The theory of labour still stands in need of further 
extension. The development of the points of view, which 
in this chapter it has been possible in most cases only to 
indicate, would probably show that there is still much to be 
harvested in this region. For we have even now an inkling 
of the truth, that in union of labour and labour in common 
much more subtle psychical influences cooperate than in 
the division of labour, which has hitherto been the almost 
exclusive object of our attention. To discover them all is, 
indeed, possibly only to the reflecting and self-observant 
worker. 



CHAPTER VIII. 

DIVISION OF LABOUR. 

In most of the sciences nowadays there are popular 
truths. They consist as a rule of general principles, to 
which their propounders have given such initial complete- 
ness of form and substance that it would seem as if they 
might be added at once to our store of knowledge as an as- 
sured acquisition of the human mind incapable of being 
either shaken or lost. Such truths become the mental 
property of the educated with a rapidity often surprising. 
The convenient impress they bear from the beginning 
makes them coins for intellectual exchange that gain cur- 
rency far beyond the department of knowledge for which 
they were issued. On the other hand, their passage over 
into the intellectual and linguistic circulation of the edu- 
cated world serves again to confirm their validity within 
the narrow department of study from which they have 
sprung. If knowledge is making rapid progress in this 
department it comes to pass that these now popular truths 
remain inviolate while all the remaining structure of the 
science is demolished and rebuilt. They are like inorganic 
bodies overrun and enveloped by the luxuriant growth of 
living organisms. 

Such is the case, if we are not mistaken, with the theory 
of division of labour in political economy. In its present 
form this theory dates from Adam Smith, and its popular- 
ity is indeed due in no small measure to the external cir- 
cumstance that it is presented in the first chapter of Book I 

882 



DiyiSION OF LABOUR. 283 

of his classical work, where it could not escape even the le- 
gion of those who merely " read at *' books. Adam Smith 
is, of course, not the originator of the theory. He borrows 
it in its essential features from the Essay on the History of 
Civil Society f which his countryman, Adam Ferguson, pub- 
lished in 1767. Yet the theory has been adopted by all 
later students in the agpreeable form in which Smith pre- 
sented it. In this form it has also gone over into other 
sciences and become familiar to every educated person. 

In essaying then to subject the economic theory of di- 
vision of labour to a critical examination, and to supple- 
ment this examination by the application that this theory 
has quite recently received in the department of sociology, 
we count upon dealing with a circle of ideas familiar to 
many.* For this last application marks at the same time 
one of the few attempts made by economic science to ad- 
vance on this point beyond Adam Smith. In other re- 
spects students have contented themselves with correcting 
Smith's theory of division of labour in subsidiary points, 
tracing it back historically and dogmatically to the ancient 
Greeks, adapting explanatory examples to the technical 
advances of the present, and besides its bright sides, bring- 
ing forward the dark sides as well. On the whole, how- 
ever, our remarks on popularized scientific truths hold 
good for this theory. While round about it the structure 
of economic theory has been diligently altered and 
extended, it has remained intact. Only recently a 
reputable economic writer, in a critical survey of the pro- 
gress of political economy since Adam Smith, stated that 
the subject is exhausted; that regarding it one can but re- 
peat what has been already said by others.^ 

^ Comp. following chapter. 

* M. Block, Le progres de la science economique depuis Adam Smith 
'(Paris, 2d ed., 1897), I, p. 533- 



a84 DIVISION OF LABOUR. 

Under these circumstances it will suffice for us to discuss 
the subject in direct connection with the celebrated Scotch- 
man's presentation of it. We will, however, not cover the 
whole subject, but attempt merely to answer the two ques- 
tions: What is division of labour? and How does it operate f 

What division of labour is we can nowhere learn from 
Adam Smith. He illustrates the process that he designates 
by this name only by individual examples, and from them 
deduces directly the statement which has since been 
termed the " law " of division of labour, and which can be 
sumniarized in the words, that in every industry the pro- 
ductivity of labour increases proportionately with the ex- 
tension of labour.' His examples, however, when closely 
scrutinized by no means illustrate similar economic 
processes. 

There is first the celebrated instance of the pin-manufac- 
tory. With the ordinary workman, who is not particularly 
adept at this special branch of business and who perhaps 
could, with his utmost industry, make scarcely one pin in 
a day, and certainly could not make twenty, Smith con- 
trasts the factory in which a considerable number of work- 
men with divided labour produce similar wares. " One 
man draws out the wire; another straights it; a third cuts 
it; a fourth points it; a fifth grinds it at the top for receiv- 
ing the head; to make the head requires two or three dis- 
tinct operations," etc. In this manner there result up to 
the completion of the pin eighteen distinct operations, each 
of which can be transferred to a particular hand. Smith 
finds that in such a cooperating group of workers the out- 
put of each individual, as compared with that of the la- 

•Thc correctness of this sharp formulation is manifest from the 
words of the first chapter: " The division of labour, so far as it can 
be introduced, occasions in every part a proportionable increase of the 
productive powers of labour." 



DiyiSION OF LABOUR. 285 

bourer working separately and producing the whole pro- 
duct, is increased a hundred, indeed a thousandfold. 

This example has been repeated even to weariness; it 
has become, in general, the classic type of division of 
labour. Most people can conceive of it only in this 
one form, the fbrm of a factory in which the total labour 
necessary to the production of the ware is divided into as 
many simple operations as possible, carried on simultane- 
ously by different persons in the same establishment.* 

But Adam Smith has not confined himself to this exam- 
ple. He calls it also division of labour when a product has 
to pass through various trades and employments in a coun- 
try, from the procuring of the raw material till it is ready 
for use; as, for instance, the wool through the hands of the 
sheep-breeder, the spinner, the weaver, and the dyer. In 
a rude state of society all this, he points out, is the work of 
one man; in every improved society, on the contrary, the 
farmer is generally nothing but a farmer; the manufacturer 
nothing but a manufacturer. The labour, too, that is 
necessary to produce any complete manufacture, is almost 
always divided among a great number of hands. 

Smith makes no distinction between the two kinds of 
division of labour, and ascribes to both the same eflfects. 
But it does not require lengthy consideration to recog- 
nise that we are here dealing with two distinct processes. 
In the case of the manufacture of woollen cloth a whole 
process in production is separated into various depart- 
ments. Each of these departments becomes an inde- 
pendent economic organism; and a ware must, from the 

*Helmolt, De laboris divisione, 1840, (a Doctor's Dissertation from 
the University of Utrecht,) pp. 38, 39: " Ubi plures operarii simul opus 
quoddam conficitvit, singuli vero continue eadem operis parte sunt 
occupati, ut, si aliquid perfecerint eandem rem de novo aggrediantur." 
And yet Fergruson had previously entitled his chapter on the division 
of labour: " On the Separation of Arts and Professions." 



286 DIVISION OF LABOUR, 

procuring of the raw material on, pass through a series of 
trades before it can be offered ready for consumption, each 
change in ownership involving a charge for profits. In the 
case of the pin, on the contrary, the manufacture of the 
object of the division of labour does not constitute a com- 
plete process in production, but merely a single depart- 
ment. For its raw material, the wire, is already well ad- 
vanced towards completion. The result of the division is 
not a series of new trades, but a chain of dependent em- 
ployments whose successful utilization under present con- 
ditions postulates the existence of wage-workers held to- 
gether by one entrepreneur. The product, before it is 
completed, passes, it is true, through a larger number of 
hands than previously, but it undergoes no change of pro- 
prietorship. 

Two industrial processes so thoroughly different de- 
mand different names. We will designate the division of a 
whole process of production into several industrially inde- 
pendent sections division of prodtiction; and the breaking 
up of a department of production into simple dependent 
labour elements subdivision of labour} 

Finally Adam Smith cites a third example that is neither 
division of production nor subdivision of labour. He com- 
pares three smiths: a common smith, who can handle the 
hammer well, but has never been accustomed to make 
nails, a second smith who has been accustomed to make 
nails, but has not this for his sole or principal occupation, 
and finally a nail-smith who has never followed any other 
calling. He finds that if all three make nails for a definite 
period the work done increases according as the workman 
limits himself to the production of one product. It is this 
limitation to the exclusive production of a single line of 
goods that he calls division of labour. 

* Produktionsteilung and ArheitszerUgung. 



DIVISION OF LABOUR. 287 

The justification for this nomenclature is not at once 
apparent. What is divided? Where are the parts? 

Manifestly Smith conceives the whole business of a 
smith who, as in olden times, makes horseshoes, plough- 
shares, and wheel-tires, as well as axes, spades, and nails, 
as the subject of the division. From this comprehensive 
department of production a line of products is separated, 
and their production taken over by a special workman, the 
nail-smith, while the remaining products continue to form 
part of the smith's work. The articles fonnerly produced 
jointly in the one business of the smith are for the future 
manufactured in two different businesses. In the place of 
one industry there are now two; and each forms for an 
individual a separate business or vocation. 

It is clear that in this case it is neither a question of cut- 
ting a somewhat extensive process of production into va- 
rious sections, nor of subdividing such a section into its 
simplest elements. For, as Smith himself explains, the 
labour process of the nailer is neither shorter nor less 
complex than the smith's: each for himself blows the bel- 
lows, stirs the fire, heats the iron, and forges every part of 
the product. A change has taken place only in one re- 
spect: each applies this process to fewer classes of goods. 
Under the system of divided labour, however, the goods 
produced, taken singly, do not pass through more hands 
than formerly. We will call this third species of division 
of labour specialization or division of trades. 

How specialization is distinguished from subdivision of 
labour is readily perceived. The one is a division of the 
whole task of production between different businesses; 
the other takes place within a single business. It is per- 
haps more difficult at first sight to distinguish division of 
production and specialization of trades. In division of 
production cross-cuts, as it were, are made through a 



2S8 DiyiSlON OF LABOUR. 

somewhat extensive process of production; in specializa- 
tion a distinct department of business is split lengthwise. 

To offer a simple example, the production of leather 
articles of use was originally confined to the one estab- 
lishment. The Siberian nomad and the Southern Slav 
peasant still procure the hides, tan them, and out of the 
leather make footwear, harness, etc., within their own 
household establishment. In the countries of Western 
Europe the trades of the tanner and the currier had 
arisen by the early Middle Ages. Leather goods 
down to their finished condition then passed through 
three trades, — that of the furnisher of the hides, of the 
tanner, and of the currier. That was division of produc- 
tion. In time the special handicrafts of the shoemaker, 
the saddler, the strap-maker, the maker of fine leather 
goods, etc., have separated themselves from the large 
industry of the currier; and each produces a particu- 
lar class of leather wares by approximately the same 
process of work. That is specialization, or division of 
trades. 

In division of production, to use a simile, the whole 
stream of production of goods is from time to time 
dammed up by weirs; with specialization it is diverted into 
numerous small channels and rivulets. 

In his explanatorj' examples Smith goes no farther 
than this. We may also for the present pause here 
and lay before ourselves the question: What led the 
" father of political economy " to embrace under the one 
name division of labour three processes so different as 
division of production, subdivision of work, and specializa- 
tion? Wherein are these processes, whose fundamental 
differences we have been able only briefly to indicate, es- 
sentially similar? 

The true response to this question will at the same time 



Diy^lSlON OF LABOUR. 289 

furnish us with the simplest and broadest definition of 
division of labour, a definition which must be accepted by 
all who, on this point, have followed Adam Smith, that is, 
by all scientific political economists.* 

Manifestly those three different kinds of division of 
labour have only the following in common with each 
other: all three are processes in the economic evolution of 
society which have originated through acts of human volition, 
and in which an economic task is transferred from the one 
person hitherto performing it to several persons, the transfer 
being so made that each of these performs but a separate pat^t 
of the previous total labour. Division of labour will ac- 
cordingly always be characterized by an increase in the 
number of labourers necessary for the accomplishment of 
a definite economic end, and at the same time by a differ- 
entiation of work. The economic tasks become simplified, 
better adapted to limited human capacities; they become, 
as it were, individualized. Hence division of labour is al- 
ways at the same time classification of labour, organiza- 
tion of labour in accord with the economic principle; its 
result is ever the cooperation of varied energies in a com- 

* Those savants of course excepted who no longer define at all. 
Most later definitions overlook the causal force of the word division, 
and in place of the process of division put the realized condition. 
Schmoller, for instance, understands by division of labour " the per- 
manent adaptation of the individual to a specialized life-work affecting 
and dominating the whole life" (Jhrb. f. Gesetzg., Verw. u. Volksw., 
XIV, 47). He thus forces under division what can be but its result. 
E. v. Philippovich states in his Grundriss d. Pol. Oek., I, 50: " Division 
of labour is the actual divided performance of tasks leading to a com- 
mon end. It assumes, like every division, a unity from whose stand- 
point the labour of t]>e individual appears not as something exclusive 
and self-contained, but as a part of a larger whole. This unity is de- 
termined either by society as a whole, or by some organization of 
society into separate, parts." But why first construct this totality? 
Why not begin with it? Society and the business undertaking have 
surely not been divided; they are but res«dts of the division of labour. 



290 DIVISION OF LABOUR. 

mon work which could formerly only be performed by a 
single pair of hands. 

Keeping this clearly in mind, and passing in review from 
this standpoint the whole field of the economic employ- 
ment of labour in its historical and contemporary develop- 
ment, we soon recognise that with the typical examples 
of Adam Smith and the three kinds of division of labour 
deduced by us from them, the range of the latter is by no 
means exhausted. We find, on the contrary, a fourth and 
a fifth type of division of labour, which we will designate 
respectively the formation of trades and the displacement of 
labour. 

Let us begin with the formation of trades , which should 
indeed rank first in an enumeration of the kinds of 
division of labour. For it forms the beginning of all eco- 
nomic development. To understand it one must start 
from the conception, that before the origin of a national 
economy the different peoples pass through a condition of 
pure private economy in which each house has to produce 
through the labour of its members all that is required. 
This labour can be divided among the members of the 
household in various ways, according to age, sex, and 
physical strength, and according also to their relation to 
the father of the family. But this distribution of labour is 
not division of labour from the standpoint of society, for its 
effects remain restricted to the household and exert no 
creative influence upon other economies; nor does it in- 
fluence the formation of classes in society. At this stage 
there are, therefore, all varieties of agricultural and in- 
dustrial technique; but there is no system of agriculture, 
no industry, no trade as a separate branch of business; 
there are no peasants, no industrial classes, no merchants 
as social business groups. 

This state of affairs is altered as soon as individual tasks 



DIVISION OF LABOUR. 291 

separate from this many-sided activity and become sub- 
jects of vocations, the bases of particular business occupa- 
tions. The way for this advance -is prepared by the 
division of labour of the great slave and serf husbandries. 
We cannot, however, treat of these here. The part that 
detaches itself from the range of work of the autonomous 
domestic establishment and becomes a separate and inde- 
pendent business is at one time a complete process of 
production, for example, pottery; or again, a single 
section in production, for example, cloth-fulling, corn- 
grinding; • or still again, a species of personal service, for 
example, surgical work. Most frequently, however, it is 
the productive part of the domestic labours that is 
abridged through the formation of trades; and in the 
course of centuries these labours are more and more re- 
stricted to the province of consumption. On the other 
hand, there arise the different branches of production and 
the various industries which through specialization and 
division of production become multiplied ad infinitum. 

It would be a mistake, however, to imagine that this 
process of the formation of vocations which begins with 
us in the early Middle Ages has long been completed. 
Parts of the old domestic economy are still falling 
away; in the country, slowly; in the towns, more rapidly. 
Every city directory can disclose to us a series of inde- 
pendent industries which have come into existence only 
within the present century, through the splitting up of 
former single phases of domestic activity. 

Of course it would be erroneous to assume that each 
instance of a new trade which is not division of an old trade 
or of a branch of production is to be traced back to divi- 
sion of labour between household and new business occu- 

*In this case the formation of a trade is at the same time division 
of production. 



29« DiyiSlON OF LABOUR. 

pations. A bicycle-factory, a galvanizing or electrical 
establishment, an ice-factory, and a photographic atelier 
are industrial undertakings owing their origin not to 
division of labour, but to the rise of entirely new species of 
goods. They must accordingly be excluded from this sur- 
vey. Yet they are not on that account beyond the influ- 
ence of division of labour, since from the beginning they 
accommodate themselves to the forms of production con- 
ditioned by such division. 

Connected only externally with this process is the phe- 
nomenon that we have already termed displacement of 
labour. It accompanies the invention of new machines and 
other fixed tools of labour. The division of labour here 
operates in the following manner: 

With the introduction into a branch of .production of a 
newly-invented machine there is a complete displacement 
of the previous organization of work. As a rule the 
mechanism undertakes only separate movements that un- 
til then have fallen to the human hand ; and in the business 
installing the machine the sole initial change is generally 
the transfer of the workman, who formerly performed 
those muscular motions, to attendance upon the machine, 
which demands from him other muscular motions. In 
this way, for instance, after the introduction of the sewing- 
machine, the labourer in the tailoring establishment works 
with hand and foot, while formerly his hand alone was 
called into action, and then, moreover, in a different man- 
ner. 

m 

But even previously to that there were many more per- 
sons engaged in the making of a coat than the tailor* 
There were in the first place the producers of the ma- 
terials used by the tailor: the wool-producer, the spinner, 
the weaver, the dyer, etc. ; then the producers of fine im- 
plements: the needle-manufacturer, the scissors-manufac- 



DIVISION OF LABOUR. 293 

turer, and many others. All these producers still continue 
active after the introduction of a sewing-machine. But a 
new one is added, the machine-manufacturer; or, since the 
machine is produced through subdivision of work, at once 
a whole group: the machine-fitter, the founder, the metal- 
turner, the carpenter for the models, the mounter, the var- 
nisher, etc. We have, if we embrace under our view the 
whole process of production, a part of the total labour 
pushed back from a later to an earlier stage. The work 
of tailoring is in part transferred from the tailoring estab- 
lishment to the machine factory. 

The whole process is typical, and undoubtedly exhibits 
the characteristics of division of labour. If we employ for 
it the expression displacement of labour the phrase must be 
understood in a local and temporal sense. As regards 
locality, displacement of labour means the partial trans- 
ference of the manufacture of an article from one place 
of production to another; as regards time, it signifies the 
substitution of work that has been previously performed 
for work being performed now, the pushing back of a sec- 
tion of the work that was formerly devoted to the pro- 
duction of consumption goods to the furnishing of the 
means of production. In this, however, it is not at all 
necessary that a new undertaking should be formed to 
produce the new implement of labour exclusively. For, 
as in the case of the sewing-machine, a machine-factory 
already in existence can undertake its production. The 
essential thing to note is that the new process for the pro- 
duction of clothing contains an increased number of dif- 
ferent employments, and accordingly claims the service of 
more labourers. 

We have now become acquainted with five different 
kinds of economic processes falling under the head of 
division of labour, which are still in operation every 



294 Dlf^ISION OF LABOUR. 

day before our eyes. This is of course saying very little as 
to their relative importance in modern industrial life. For 
the latter is itself the result of a long process of develop- 
ment; and he who regards it with the eye of an historical 
student will find everywhere side by side the most primi- 
tive and the most modem: the one with a modest, the 
other with a ubiquitously prominent sphere of influence. 
Society in its long evolution from the isolated to the social 
economy has ever been seeking and finding new methods 
of organization in work. But it has not on that account 
discarded the old, nor will it discard them so long as their 
roles have not been completely played. For here too the 
great law of economy prevails; nothing is lost that is still 
capable of being at any point advantageously employed. 

This also holds for the various forms of division of 
labour. Even though subdivision of work and displace- 
ment of labour at present surpass in importance specializa- 
tion and division of production, and even though forma- 
tion of trades as a species of division of labour need hardly 
be longer taken into account, none of these principles of 
economic organization has ceased to operate. Each con- 
tinues active in the places where it can still assert its force. 

In economic history each of them has had a period of 
preponderance. Formation of trades appears with us in 
the early Middle Ages. The chief activity of specializa- 
tion is coincident with the prime of municipal develop- 
ment. Division of production begins at the same time. 
Its whole force in the economy of capital, however, is de- 
veloped only after subdivision of work and displacement 
of labour begin to operate, and neither of these can with 
certainty be traced back beyond the seventeenth century. 

It is with reluctance that we refrain from a detailed 
presentation of the historical conditions governing them, 
as well as of the causes and consequences of their appear- 



DiyiSION OF LABOUR. 295 

ance. We are the more loth because only in this way can 
the sharp distinctions we have made between the different 
processes find their full justification, and the traditional 
abstract treatment of the whole matter its refutation. We 
must, however, devote a few general words to the cause 
and result of division of labour. For the distinguishing of 
those five kinds of the latter would necessarily appear 
scientifically unimportant, or an idle nicety of refinement, 
if all stood in like casual connection with the economic 
phenomena that precede or follow them. 

Adam Smith derives all division of labour from one com- 
mon origin: man's natural propensity to trade. He does 
not determine whether this is the result of instinct or of 
conscious mental action. He thus renounces a sharp psy- 
chological analysis of economic action, and contents him- 
self with considering division of labour as deep-rooted in 
the dark depths of instinct. 

In this, however, he falls foul of his own examples. For 
if division of labour has its origin in an immemorial in- 
stinct of man, then it is a fundamental factor of economic 
life, which must assert itself whenever and wherever men 
exist. Yet Adam Smith's examples regularly set over 
against the condition of divided labour a condition of un- 
divided labour, and deduce the former from the latter. 
For this has to be inferred from the dynamic employment 
of the word division. There actually existed for centuries, 
as we already know, a condition of society in which divi- 
sion of labour was wanting: and the different kinds of the 
latter can be pretty definitely determined by the time of 
their origin. Social division of labour is thus generally a 
historical category, and not an elemental economic phe- 
nomenon. 

The same is true of exchange. Just as there have been 
epochs without economic division of labour, so there have 



29^ DIVISION OF LABOUR. 

been epochs without exchange. The first acts of trade do 
not appear simultaneously with division of labour, but 
long precede it. They serve the purpose of equalizing 
casual surpluses and deficiencies that have made their ap- 
pearance in otherwise autonomous economies. Exchange 
is here something accidental; it is not a necessary con- 
comitant of the husbandry of the time. Even when with 
the formation of trades social division of labour arises, ex- 
change is still very active in forms which it is the evident 
purpose to exclude as far as possible. The housewife of 
early time uses the hand-mill to grind the corn she herself 
has grown, and from the flour thus produced she bakes 
her bread. After the industries of the miller and the baker 
have been established the grinding is turned over to the 
miller, and the baker then receives the flour to make 
bread of. From raw material to finished product the new 
article of consumption has never changed its proprietor. 
For their pains, miller and baker are allowed to retain a 
definite part of their product. In the whole process of 
production with divided labour this is the sole occurrence 
partaking of the nature of exchange. 

From this one readily recognises that the alleged pro- 
pensity to trade is for Adam Smith only a means of extri- 
cating himself from an embarrassment. We can the more 
readily spare ourselves the trouble of entering further into 
this point since recent economists have not accepted this 
tenet of their Scottish master. They rather prefer to re- 
gard exchange as the unintentional result of division of 
labour. This we can accept, with the limitation that with 
divided labour exchange becomes necessary from the mo- 
ment that the producer possesses all the means of produc- 
tion. It then becomes a vital element of each economy; 
and from this point on almost every advance in division of 
labour increases the number of necessary acts of exchange. 



DIVISION OF LABOUR. 297 

But this stage of development is not reached till centuries 
after the earliest origin of economic division of labour. 
Even to-day, for example, it is by no means the rule in 
country parts for the miller to own the corn and the baker 
the flour, and a triple exchange to be necessary before the 
consumers can come into possession of the bread. 

If then exchange is merely a secondary phenomenon in 
the evolutionary processes of social division of labour, we 
are by this very fact compelled to seek another motive 
for man's efforts toward this division. 

In this we are led back directly to the fundamental facts 
of economy: the boundless extent of human needs, and 
the limited means of satisfying them. Human needs are 
capable of an infinite multiplication and subdivision ; they 
are never at rest; they increase in degree and extent with 
the progress of civilization. The material suitable for 
human ends is limited, as is also human labour which in- 
vests it with the qualities of a marketable ware and in- 
creases its quantity. With the increase in the number of 
human beings the relation of the total demand to the 
mass of raw material capable of profitable utilization 
which nature can offer becomes even more unfavourable. 
The quantity of labour necessary to satisfy the total re- 
quirement thus increases for a double reason: more and 
better goods are to be produced; and they are to be 
turned out under more unfavorable conditions. The share 
in head-work falling to each one engaged in the under- 
taking would thus become at length intolerable were it not 
possible to reduce it through an economic employment of 
labour. 

Now simple observation teaches that every person is 
not equally qualified by nature for every employment. 
The differing bodily and intellectual tendencies of indi- 
viduals njecessarily ocpsion important differences in the 



298 DIVISION OF LABOUR. 

products of labour; and these differences become ever 
more marked with advancing social development or, what 
is the same, with increasing variety of work to be per- 
formed. The principle of economy requires that every- 
one's employment befit one's capabilities; for only in this 
way can labour yield its highest service. To have " the 
right man in the right place " becomes all the easier as the 
tasks increase in number and each is permanently con- 
signed to a special hand. 

Along with the multiplication of occupations comes a 
simplification. Every composite work means for the indi- 
vidual executing it a frequent change of motions, and 
every such change a waste of energy. For passing 
from one kind of movement to another calls for mental, 
and bodily accommodation to the n^w class of work, which 
means an outlay of strength yielding in itself no useful 
return. With muscular movements pursuing a uniform 
course, however, the mind's share in the work can be 
eliminated, and an automatic performance of those 
movements soon enters, which, with increasing practice, 
removes farther and farther the limits of fatigue. At the 
same time the intensity of automatic labour can be greatly 
increased, so that not only can the movements be 
continued longer, but a larger number of them is pos- 
sible within a given unit of time. An extraordinary ad- 
vance in the effectiveness of labour is the result.^ 

All this makes it, as it were, a command of economy to 
narrow the labour tasks if we are to utilize every kind of 
endowment, and avoid every bootless waste of strength. In 
most processes of production, however, we find decidedly 
heterogeneous employments united: hand-work and head- 
work; operations demanding great muscular power along 

* More in detail in Arbeit u, Rhythmus, pp. 24 ff. 



DIVISION OF LABOUR. 299 

with those in which suppleness of the finger, delicacy of 
touch, keenness of sight come in question; tasks requir- 
ing a skill gained through theory and practice, and those 
that even th% unpracticed is in a position to undertake. 
In early times when these different tasks were placed in 
one hand a great waste of skilled labour resulted, and the 
productive part of the population was limited to those 
who h?id mastered some one branch of technique in all its 
parts. By separating the qualitatively unequal labour ele- 
ments from one another, division of labour succeeds in 
utilizing the weakest as well as the strongest workers, and 
in inciting them to the development of the highest special 
skill. 

Thus division of labour is, in the last analysis, nothing 
but one of those processes of adaptation that play so 
great a part in the evolutionary history of the whole in- 
habited world: adaptation of the tasks of labour to the 
variety of human powers, adaptation of individual powers 
to the tasks to be performed, continued differentiation 
of the one and of the other. Therewith the whole 
process advances out of the twilight of instinctive life into 
the bright day of conscious human activity. 

Yet one fact still requires special mention. It is, that 
the personal casual element in division of labour stands out 
more clearly the further back we go in the history of man- 
kind. For this reason the predominant forms of division 
of labour in the early stages of development are those in 
which the individual is assigned an independent task that 
can be carried out without any extensive material equip- 
ment. It is more especially the intellectual and artistic 
activities that expand earliest into vocations. The priest, 
the prophet, the magician, the singer, and the dancer are 
the first to g^in a separate position on account of special 
gifts. 



300 DIVISION OF LABOUR. 

If an unfree system of labour exists, division of labour 
develops first within the slave family; and it is with the as- 
sistance of a personal and moral feature hitherto hardly 
heeded that it comes into being. Wherever the system of 
supervised labour in common is inapplicable, the master 
must provide every unfree worker with a particular range 
of duties, for which he can be held responsible. He must 
impose on him a single definite kind of work if he wishes 
to profit by his labour. Hence among the Romans that 
almost over-refined specialization of work in the fatnilia 
urbana,^ the careful selection of slaves according to bodily 
and mental endowment for the different agricultural em- 
ployments; ® and among the serfs of the Middle Ages the 
extreme frequency with which the rent paid in kind was 
fixed in very special products of domestic work.^^ The 
man acting in the slave household exclusively as field- 
worker or smith, barber or scribe, the rent-collector and 
the man whose sole duty it was to supply the court with 
casks or vessels, knives or linen cloth, acquired a special 
dexterity wherewith, when the hour of emancipation 
sounded, they entered society as professional workers. 
Thus in the individual task rendered necessary by the slave 
economy of the stage of exclusive domestic husbandry, 
and in the specialization that it conditions, lay the seed- 
time for the social division of labour of the following stage. 

It is at a much later date that material elements supple- 
ment the personal element of endowment and adaptation 

• Compare above, p. 99. 

'Compare on this the fine remarks of Columella, I, 9: "Sed et 
illud censeo, ne confundantur opera familiae, sec ut omnes omnia 
exsequantur; nam id minime conducit agricolae, sen quia nemo suum 
proprium aliquod esse opus credit, seu quia, cum enisus est, non suo 
sed communi officio proficit ideoque labori multum se substrahit; nee 
tamen viritim malefactum deprehenditur, quod fit a multis,'* etc. 

"A list is given above, on page 104. 



DIVISION OF LABOUR. 30 1 

in originating division of labour." Formerly it was men 
alone, now things also become differentiated, — tools, raw 
materials, products. Each advance in division of labour 
seeks to adapt itself to the existing tools and implements, 
or to provide new ones for the particular task. Let one 
but think of the innumerable kinds of hammers, tongs, 
and chisels used in the different branches of metal- and 
wood-working! The division of labour among persons 
finds its counterpart in a division of use among the instru- 
ments of work. But as long as the tool is merely a rein- 
forcement of the human agent, the personal adjustment 
will dominate the process of production. It is only when 
artificial appliances are introduced which enable man to 
subdue natural powers to his service that the labour in- 
strument gains control over the labourer's social indi- 
viduality, as well as over his bodily movements. And now 
the impetus for a fresh advance along the path of division 
of labour can as readily originate in a newly-invented im- 
plement of labour as in the possession or acquisition of a 
particular personal qualification. Most newly-invented 
machines require the attendance of workmen possessing a 
qualification not previously represented in the business. 
Then, joined to this, comes the saving consequent upon 
the growing extension of production, a feature of impor- 
tance from the standpoint of capital. But this saving can 
take place only on the assumption of a unification and con- 
centration of demand sufficient to make the wholesale 

" On what follows compare Arbeit u. Rhythmus, Chap. IX. How 
strongly the personal element still predominated in the division of 
labour in the town economy of the Middle Ages is seen from the con- 
ditions for admission into a guild. As far as the carrying on of a trade 
came in question only personal requirements were made — ability to do 
the work with one's own hand. Material requirements had to !>e met 
by the person seeking admission only as a citizen — possession of a 
house and of arms; and as a Christian^utrance«fee in wax. 



302 DIVISION OF LABOUR. 

production, which perhaps has long been technically pos- 
sible, economically possible also. Many labour-processes 
cause approximately the same costs whether they embrace 
many or few pieces, as is the case, for example, with dye- 
ing, grinding, drying, postal delivery. But if the method 
of work can be so contrived that masses of the raw or half- 
manufactured material which are to be worked over are 
collected at definite points, the employment of hands at 
these points solely for this purpose becomes profitable, 
with, on the whole, a considerable saving in costs. 

How far in such matters the social principles of immo- 
bility of labour and of free competition may cooperate to 
retard or to advance is not to be investigated here. A 
warning is merely to be given against observing and judg- 
ing these matters solely in the light of modem industrial 
conditions. Division of labour reaches out far beyond the 
sphere of material things. It can show in recent times,, 
especially in the field of intellectual work, advances and 
results far surpassing those in the department of manu- 
facturing technique. Indeed, the former are largely the 
direct cause and occasion of the latter. On the other 
hand, in the whole broad field lying beyond the limits of 
material production the material aids to labour play only 
an unimportant part. The personal element is here con- 
tinually decisive for the further development of division 
of labour; and we thus have to recognise it as paramount 
in the whole great process of advancing civilization. 

As to the universal originating cause of division of 
labour more than this cannot be said. The particular 
conditions of origin under which the various kinds and 
forms make their appearance will be briefly discussed in 
another place. 

At this point we can make but like cursory reference 
also to the economic consequences of division of labour, al- 



DIk'ISION OF LABOUR. 3^3 

though it is in this very particular that the various forms 
most widely diverge. 

Adam Smith knows but one effect of division of labour: 
the increased productivity of labour. He thus restricts 
its influence to the sphere of production. In this he is 
right. Division of labour permits the production of 
more and better goods with a given expenditure of hu- 
man strength than was pK)ssible with undivided labour. 
Production becomes cheaper; its costs diminish as far as 
labour is concerned. And since Smith considers labour 
the true measure of exchange value, he can dispense with 
investigating whether under all circumstances division of 
labour also insures a cheaper satisfaction of the wants of 
the customer. 

However narrow this conception may appear, it is cer- 
tainly more reasonable than the unlimited extension given 
by many recent economists ^^ to the effects of division of 
labour when they derive the whole of our present com- 
mercial organization directly from division of labour, and 
think to characterize it sufficiently by calling it, as they 
commonly do, the " economy of divided labour." In this 
they allow themselves to be guided by the opinion that in 
their present form and method of action the most im- 
portant economic phenomena are determined by division 
of labour; that in the highly refined subdivision of trades 
occasioned by it division of labour is, so to speak, the 
skeleton supporting the economic organism, while trade 
and commerce represent the ligaments and muscles that 
hold it together and enable it to functionate like a great 
living body. Commerce, however, they say, is occasioned 
directly by division of labour; division of labour is its 
cause. 

" Schmoller may again serve as an example: Grundriss, I, pp. 364(1 



304 DIVISION OF L/iBOUR. 

Therein lies a great mistake. By itself, division of 
labour does not create trade. And inversely, a condition 
of undivided labour may easily be imagined concurrent 
with a relatively well-developed trade. 

Let us elucidate the last sentence first. We may recall 
that peoples standing at the stage of private domestic 
economy can have a relatively well-developed exchange 
of goods — for profit or otherwise, if smallness of the 
household membership or extraordinary inequality in the 
distribution of the gifts of nature give occasion for it. 
Each house and each worker produces, in a condition of 
complete union of labour, everything that the natural ad- 
vantages of the place of habitation permit of. Exchange 
but fills up the gips of home-production; its objects are 
merely surpluses of otherwise autonomous establishments. 
The weaker the different households are numerically and 
the oftener unfavourable seasons — dying of cattle, spoiling 
of the stores, or sickness of members of the household — 
threaten at particular points the satisfaction of their needs, 
the more frequently will surplus ware be drawn from an 
outside source in exchange for the excess commodity in 
one's own sphere. 

Thus the negro races of Central Africa have a great 
number of weekly markets, which are usually held under 
special peace protection in the midst of the primaeval 
forest. Yet among them there is scarcely a single in- 
dustry carried on as a business; and every species of di- 
vision of labour is lacking, save the separation of the 
spheres of work according to sex. The same state of 
things has been observed in different parts of Oceania. 
Even in the countries of Western Europe a fairly 
brisk market trade must seemingly be assumed for the 
early Middle Ages, notwithstanding the complete absence 
of a developed subdivision of labour. 



DIVISION OF LABOUR. S^S 

On the other hand, as already frequently remarked, 
when the existence of slavery or serfdom calls into being 
households numerically extensive, division of labour 
can establish itself at the same stage of domestic work 
without giving rise to exchange. On the estates of 
wealthy Romans there were workmen of very different 
grades of skill, perhaps even some who produced accord- 
ing to the principle of subdivision of work; but exchange 
neither united them with each other nor with the con- 
sumers of their products. The power holding them to- 
gether was the authority of the head of the family. Under 
slavery this power lay in the ownership of the persons, 
under serfdom, in the ownership of the soil. An establish- 
ment thus organized is a permanent community for prodnc- 
tion and consumption. What it produces it also consumes. 
Indeed, division of labour really appears for it a welcome 
means of avoiding exchange. 

In such large households regular division of labour ac- 
cording to employment paves the way for the succeeding 
economic stage. It is the starting-point for the formation 
of trades. On the latter is based the origin of special 
economic life-vocations. It liberates a section of humanity 
from the soil, on the possession of which its existence had 
solely depended. It furnishes the burgher as well as the 
peasant with the means of livelihood. -Specialization in- 
creases the number of opportunities for trade, it supplies 
the framework within which higher mechanical skill is 
developed. And at first division of production has no 
other effect. Formation of trades, specialization, and 
division of production — ^all three together — ^are indeed 
able of themselves to create an economy based upon di- 
vided labour, but this economy is not at once national 
economy. For, first of all, it still lacks the circulation 
of goods. 



3^6 DIVISION OF LyfBOUR. 

The whole process of division of labour up to this point 
proceeds, as we know, by the method of workers separat- 
ing from the independent household of the proprietor of 
the soil, and in the service of other households turning to 
account any special skill in the form of wage-work. They 
are, it is true, tradesmen living from the earnings of their 
special trade; but the raw material that they work up is 
owned by the person who will finally consume the product 
in his own house. Now there are certain cases in which 
several of such wage-workers must cooperate in one 
process of manufacture if the commodity is to be com- 
pletely finished; for example, in preparing bread, the 
miller and the baker; in making a garment, the weaver, 
the dyer, and the tailor. In the exercise of their technical 
skill all these labourers engaged in independent trades 
are united with one another through the product which 
passes through their hands in different stages of its manu- 
facture. The whole employment of the one is to con- 
tinue the work of the other. Their economic coopera- 
tion, however, is effected by the owner of the raw ma- 
terial, who has himself generally produced it, and to whom 
the finished product returns — that is, by the consumer. 
The means, however, by which the same person attracts 
to his service the various part-producers is the wage that 
he pays to each. This payment, moreover, represents the 
sole commercial act involved in this kind of division of 
labour. 

In building a house one employs successively on wage 
the mason, the carpenter, the roofer, the glazier, the 
joiner, the locksmith, and the decorators, and supplies 
them with the material necessary for their work. Their 
objective point of union they all find in the new structure; 
their personal point of union they possess in the builder. 
He unites them, so to speak, into a temporary community 



DIVISION OF LABOUR. 307 

of production. But their union is a loose and constantly 
changing one. No permanent economic organization of 
society arises from it. To-day they serve this builder, to- 
morrow that. Division of labour makes the producers so- 
cially dependent neither upon one another nor upon the 
contractor. They remain " master workmen." 

Nor is there much change in this regard when the wage- 
worker rises to the position of craftsman and himself sup- 
plies the raw material for his labour. A wagon, for in- 
stance, is ordered from the wagon-maker, is ironed by the 
smith, and painted by the painter. The wagon-maker fur- 
nishes the wood, the smith the iron, the painter the paint. 
The payment that they receive at this stage is a remunera- 
tion for the labour and the material furnished by each. 
But the one guiding the production is still, as ever, the 
consumer of the commodity produced by divided labour. 

Through all earlier forms of division of labour, as one 
perceives, there runs an obvious endeavour to restrict the 
number of commercial transactions evoked by it to 
those absolutely necessary. In the midst of all trades 
originating in division of labour stands domestic work, 
the mother of them all, with its primeval community of 
labour dissolving but slowly. Alongside it, even through- 
out the stage of town economy, the particular manufac- 
turing establishments and professional workmen called 
forth by formation of trades, specialization, and di- 
vision of production continue to be firmly and closely 
united. From the customer's house they receive the com- 
missions which they execute; and even then during the 
performance of the work they frequently enter into a tem- 
porary consuming community}^ 

In the stage of national economy the consumer with- 

" [Comp. Chap. IV, remarks on iVin^ancy.— Ed.] 



ZoB DIVISION OF LABOUR. 

draws more and more from his century-old position as 
director and uniter of divided production. These duties 
now develop into a vocation. This vocation, however, 
can be independently exercised by those alone in whose 
hands the means of production — or at least the circulating 
means of production — are at the same time found, that is, 
by the capitalists. Because of the double duties that thus 
fall to them — procuring capital and directing the produc- 
tion — they are called business undertakers or entrepre- 
neurs. 

In their hands the division of labour undergoes a com- 
plete transformation. In so far as it is division of produc- 
tion each part-producer now disposes of the products of 
his own raw material to his successor. They become for 
each a source of profit, that is, circulating capital. Thus 
arises along with the trade in certain classes of finished 
wares, a series of exchanges of raw material or unfinished 
goods with no other aim than to unite the various stages 
of division of labour with each other. This exchanging 
is in character quite unlike the one between the consumer 
and the various producers in succession, which previously 
held exclusive possession of the field. The earlier ex- 
change, at least for the one acquiring the product, is pure 
exchange for use, in which he is concerned with the com- 
modity as an object of consumption. The later exchange is 
for purchaser and seller always a business transaction in 
which the utility of the object of exchange is of secondary, 
and its character as capital — the profit to be gained by it — 
of primary, importance. The forms of division of labour, 
displacement of labour, and subdivision of work now aris- 
ing for the first time have by their mutual relationship the 
effect of imparting the quality of capital to the fixed means 
of production as well. The subdivision of work makes 
necessary a permanently dependent labouring class. It 



DIVISION OF L/IBOUR. 309 

alone gives to the method of capitalistic production its 
full expansion. Although, on the other hand, it largejy de- 
stroys, in the department accessible to it, that which the 
formation of trades and specialization had previously cre- 
ated — the independence of the petty traders. 

This new phase of division of labour, accordingly, raises 
commerce to a heig*ht unknown before. In trade, in trans- 
portation, in credit negotiation, in insurance, it calls forth 
numberless other phenomena of division of labour under 
the shadow of the entrepreneur system, which lead in turn 
to fresh commercial services of a manifold kind. But in 
itself division of labour does not create this new com- 
merce. The impelling and creative element in modem 
national economy is not division of labour, but business 
capital, and commerce is its spring of life. 

The point at which capital in its primal form of money 
first displayed its earning power was trade. From there 
it has encroached upon production by enabling the trader 
to take the consumer's place as director of production. 
In this way that commission system first took its rise in 
the world of industry in which the commission manufac- 
turer enters into the same outer relationship with wage- 
worker and craftsman that the father of the household 
formerly held. To the one he advances the raw material, 
from the other he purchases the finished products made 
from self-supplied material, with the object of further dis- 
posing of them. Where a productive process falls into 
different sections he guides the product from one to the 
other, and finally places it on the market as finished ware. 
As a rule, he operates merely with circulating capital. 
He has to do permanently with fixed capital only when it 
becomes profitable to pass over from commission to fac- 
tory production. 'While, however, in the province of 
industry trading capital was merely a transforming agency. 



3IO DimSlON OF LABOUR. 

in the departments of banking, transportation, and in- 
surance it has been creative. These departments of busi- 
ness are reaHy,- when we consider them from the side of 
division of labour, only ramifications of trade. 

Thus, it seems to us, we have to recognise capital as the 
creative inRuence in modem national economy, and division 
of labour as the medium through which it operates. Its 
support and representative is the entrepreneur. 

That the latter has been able to utilize this medium of 
division of labour to much greater purpose than the head 
of the household before him is plainly evident. To-day the 
entrepreneur determines what we shall eat and drink, read 
in the papers and see at the theatre, how we shall lodge 
and dress. That means everything. For a great part of 
the goods we consume the right of self-determining is 
taken away. And since uniform production on a large 
scale is most advantageous to the manufacturer, there is 
operative in the sphere of consumption an increasingly 
active uniforming process. 

In contrast with this the province of labour exhibits a 
continually advancing differentiation. The field of work 
of each individual is ever growing more restricted. It is 
only when broken up on the basis of technique into 
its parts, that labouring skill can furnish workable building 
material for the task of the entrepreneur. Every business 
establishment is a union of various fragmentary activities, 
originating through division of labour, into an organic 
whole. It unites workmen economically and technically 
dependent into a permanent community of production. This 
community of production, however, has ceased to be at the 
same time a community of consumption. On the contrary, its 
members belong to distinct households which have been 
freed from all the burdens of production, and which are in 



DIVISION OF LABOUR, 311 

no wise connected with one another or with the employer's 
household. 

In the formation of those communities of production 
the entrepreneur's plan of action varies according to the 
presence or absence of the earlier forms of division of 
labour in the manufacture in which he wishes to place his 
capital. 

In the first case he absorbs into his undertaking all the 
independent branches of business that up to that time had 
to do with the wares to be produced. He specializes their 
workers and permanently allots to them the performance, 
side by side, of the part-tasks required by the business. As 
an example, take the furniture-factory, in which joiners, 
turners, wood-carvers, upholsterers, glaziers, painters, 
and finishers are incorporated in a common productive 
process. 

In the second case he first organizes the work according 
to the principles of subdivision of labour, in the branch of 
production concerned, and furnishes the business with a 
comprehensive outfit of machinery. 

In both instances there are in the fully-equipped busi- 
ness, in addition to the entrepreneur, only subject part- 
workmen whom the technical arrangement of the work 
renders dependent. In the one case they have been inde- 
pendent craftsmen, and the task of the entrepreneur con- 
sists in combining them into one industrial unit; in the 
other the business unit already exists, and its component 
parts are to be sought. Very soon the employees of 
either origin are no longer to be distinguished from one 
another. 

Early handicraft had as a basis a few workmen of 
similar training who, even though at different stages (ap- 
prentices, journeymen, master workmen), worked side by 
side. The qualifications of the groups so composed bear. 



3" DIVISION OF LABOUR, 

from handicraft to handicraft, no resemblance to each 
other. It is impossible for a transfer to be made from one 
species of employment to another; for instance, the smith 
cannot be wheelwright. The law recognises this by the 
sharp dividing lines it draws between them. 

Modern industrial activity unites workers differing in 
skill and strength into cooperative harmony within the 
undertaking. Their grouping for business purposes fol- 
lows the same principles of organization from branch to 
branch of production: there are no sharp boundary lines 
between industries. A distinction of vocations hardly oc- 
curs among entrepreneurs, though to a certain extent it 
exists among the workmen. As far as the entrepreneur's 
functions are concerned, it is almost a matter of indiffer- 
ence whether he manages a street-railway, iron-works, 
or a weaving factory. Among the employees, on the con- 
trary, in consequence of the continued subdivision of work, 
there are now numerous specialists who are required in 
very different branches of production. The locksmith, the 
metal-turner, the moulder, the planer, and the cutter ap- 
pear in all branches of well-advanced metal industry, in 
each special department of machine-construction, in rail- 
way workshops, etc. Fireman and engineer are neces- 
sary in every large establishment, whether it produces 
cotton thread or illustrated papers. Joiners, tinsmiths, 
coopers can be incorporated into or attached to under- 
takings of the most varied type, and office-clerks, pattern- 
artists, and engineers have a similarly varied usefulness. 
To these is to be added the mass of unskilled labour that 
is swallowed up by the large manufacturing establish- 
ments. For many entrepreneurs almost the sole remain- 
ing question is how to apportion and arrange these labour 
elements in such a way that they may cooperate as a me- 
chanical unit. 



DIVISION Of LABOUR. 313 

This cursory survey has taught us how at different 
epochs division of labour has exerted an influence upon 
the industry of peoples, and upon the existence of in- 
dividuals, varying according to the organizing principles 
dt)minating the different economic stages. 

During the stage of independent household economy 
there prevails either union of labour in the hands of the 
father and mother of the family, or division of labour de- 
veloped upon the basis of slavery or serfdom. In both 
cases the household represents a permanent community 
of production and consumption. The principle holds 
good: who works with me shall eat with me. 

In the stage of town economy specialization and division 
of production predominate. The part-producers are per- ^ 
sonally free; but the consumer of their wares, who unites 
them under favourable circumstances into temporary com- 
munities of production, determines in the main the nature 
and time of their production. During the period of com- 
mon production he often provides them with their keep. 

During the stage of developed national economy the 
entrepreneur controls the production of wares under di- 
vision of labour. The part-producers are personally free 
labourers. They are united by the employer into per- 
manent communities of production. All other community 
of living is excluded; and if perchance on occasion of a 
business jubilee the entrepreneur gives a dinner to his, 
workmen, the newspapers report how he ate and drank 
with them at the same table, and consider it a particular 
condescension on his part. 

These are different economic worlds, separated from 
one another by a deep gulf. If there breathes in the prim- 
itive union of labour of the home, and, in part, in the 
earlier division of labour as well, a warm breath of social 
fellowship, there surge through the modern division of 



314 DIVISION OF LABOUR. 

labour the cold, cutting winds of calculation, contract, and 
greed of gain. If the older division of labour was the 
caryatid of economic independence, the modem is ever 
forcing large masses into a condition of dependence. The 
pressure of capital is making men's occupations increas- 
ingly dissimilar; it is making the men, as consumers, ever 
more alike. If in the olden time the individual's portion 
of goods was shaped by his own hands and head, and was, 
so to say, a component part of his being that had taken 
objective form, the consumption goods surrounding us 
to-day are the work of many hands and heads. As to 
the workers, we are supremely indifferent; and as to their 
work, when once we have paid its last possessor the mar- 
ket price, we for the most part reck but little. In the 
narrow circle of a life-vocation the mind becomes nar- 
rowed, frequently to obtuseness. In our sphere of activ- 
ity we have lost in fulness of life, and the worker has not 
the old joy in his work. Are we sufficiently compensated 
for these losses by the variety of articles which it is ours 
to use because thousands of hands labour for us, because 
thousands of heads think for us? Or has division of labour 
merely made life richer in pleasures, but poorer in real 
joy? 



CHAPTER IX. 

ORGANIZATION OF WORK AND THE FORMATION OF SOCIAL 

CLASSES. 

The economic processes involved in the organization 
of work are processes of adaptation. Whether they fall 
under the head of union of labour, of labour in common, 
or of division of labour,* they all originate in the effort to 
remove the disproportion perchance existing between the 
labour to be performed at a given moment and the powers 
of the individual labourers, and to bring them into agree- 
ment with each other. They must accordingly react upon 
the individual in compelling him to adapt, to accommo- 
date himself mentally and physically to a definite work. 
In this adaptation certain resistances on the part of human 
nature are first to be overcome. Once vanquished, how- 
ever, this negative element is, usually by virtue of con- 
tinued practice, replaced by a positive one. The individual 

' It will assist to an understanding of the present and the two pre- 
ceding chapters if we present here the various kinds and varieties of 
labour organization in tabular survey. [Comp. Chapters VII and 
VIII.— Ea] 



A. Union of Labour. 

B. Labour in Common. 



I. Fraternal labour. f ^^^ Simple aggre- 

^ ^In^lbf^r^"'"*"- 1 (^) Concktenation 
3. Joint labour. [ ^^^ j^^^^^^ 

I. Formation of trades. 

3. Specialization. 
C. Division of Labour. { 3. Division of production. 

4. Subdivision of work. 

5. Displacement of labour. 

3x5 



3x6 ORGANIZATION OF fVORK AND 

gains insight into the special character of his work; he 
develops a particular dexterity for it; his mental powers 
are directed continuously towards the same goal, and 
therefore expand in a definite direction; in short, his 
adaptation to the work becomes a part of his being and 
distinguishes him from other individuals. 

If, then, the class of work to which the workman de- 
votes himself be of such a nature as to accentuate the 
special character of the individual man in society, the 
question naturally arises, how far such individual charac- 
teristics arising from work react upon the social life of the 
species. More specifically stated, the question would be: 
Is there a definite organization of society corresponding 
to a definite organization of work; and what is the nature 
of the influence exerted by the one upon the other? 

The question is not so simple as it may perhaps at first 
sight appear. Nothing, for instance, seems easier than to 
trace back the caste system of India to the hereditary 
character of occupations, and accordingly to seek its 
origin in division of labour. But we know positively that 
the lower and the higher castes have different origins; 
and many indications favour the view that place of resi- 
dence and possession of property have cooperated in the 
genesis of that hereditary stratification of society. Finally 
we see that the essential nature of the caste lay in purity 
of blood and of social relationships. Difference in caste 
excluded eating in common especially, although it does 
not seem to have prevented a similarity of occupation. 
All this gives good ground for the assumption that the 
separation according to employments was only a result of 
the division into castes which had originated in differences 
in race.^ A similar course of development can be shown 
for the social classes of the Middle Ages. 

•Possibly the remarks made on pp. 54 ff. regarding tribal indus- 
tries give us the right cue. 



THE FORMATION OF SOCIAL CLASSES. 3^7 

In considering the relations between economic activity 
and society generally, it is never to be forgotten that they 
are reciprocal, and that with them we can seldom deter- 
mine with certainty action and reaction. Just as a 
particular kind of organization of work, when it lays 
hold of the individual for life, furnishes specially dif- 
ferentiated men to society, so society on the other hand 
has from its stratifications and its individuals to provide 
the plastic material used by organization of work. Cer- 
tain strata of society will favour distinct forms of labour 
in common and division of labour, others will place ob- 
stacles in the way of their operation. Slavery, for in- 
stance, encourages the concatenation of labour; the pres- 
ence of a numerous class of unpropertied wage-workers 
promotes subdivision of work. But those social influences 
alone are not able to produce these results; others of a 
technical and a general civilizing nature must be assumed, 
for instance, with subdivided labour, a highly specialized 
equipment of instruments of production. 

All these relationships are thus of an extraordinarily in- 
tricate nature and demand the most circumspect treat- 
ment. As a rule we can tell what features in the economic 
and social world are found side by side, but it is seldom 
that we can determine how they are mutually connected. 
In attempting, then, to discover the social bearings of 
organization of work in its various forms, we enter a field 
as yet little investigated, in which each step aside from 
the path leads into an impenetrable thicket of confused 
ideas. 

At first the oldest system of organization of work, 
union of labour, seems to have been socially unimportant. 
Its earliest appearance reaches back into the pre-economic 
period where the individual has to perform all the labour 
necessary to his maintenance. It is to be found more ex- 



31 8 ORGANIZATION OF IVORK AND 

tensively, then, in the earlier stages of independent do- 
mestic economy. The tools are simple and few, each 
must serve the most varied purposes, and everyone must 
be acquainted with their use. From work of such a type 
the impulse to a division of society, to a formation of re- 
lations of social dependence, manifestly cannot come. 
Society, it appears, must consist of a uniform mass of indi- 
vidual households; and such is its actual constitution as 
long as collective ownership of the soil prevails. Within 
the individual households, on the other hand, a separation 
of male and female work can take place. But this is not 
transferred to society; each household is in this respect 
an exact replica of the other. If social differences never- 
theless exist, their cause is to be sought in other con- 
ditions. 

Union of labour maintains this [socially trivial] charac- 
ter in the higher stages of development even up to the 
highest. To-day it is met with almost exclusively in the 
humbler spheres of economic life and in the lower strata 
of society. Here it arises in most cases from the striving 
for independence ; it is the support, the stay, and the com- 
fort of the common folk. Indeed, it can appear here even 
as recoil from an excessive division of labour.' If it were 
the sole active factor in the economic life of a people, it 
would lead to a society of lifeless uniformity and render 
a successful struggle from the lower to the higher im- 
possible. 

With labour in common it is different. To be sure, in its 
loosest form of fraternal labour it exists between equals 
only temporarily, and therefore can have scarcely any 
effect upon the organization of society. At the most, it 
can but suggest it. The two forms of labour aggrega- 

• Comp. our remarks in the Handwort. d. Staatswiss., IV, p. 377. 



THE FORMATION OF SOOAL CLASSES, 319 

tion, on the contrary, become a means to the formation 
of special groups; they create and maintain relations of 
social dependence or, at least, assure their continuance 
where they have been developed from other causes. The 
same can be said, although not with equal definiteness, of 
many forms of union of labour. In both cases cooperation 
amongst a plurality of persons depends upon the extent 
of the work to be performed as compared with the im- 
perfect nature of the tools; and where those tasks are of 
a permanent nature or, at least, are frequently repeated 
in any one department of economic labour — for example, 
in agriculture — they require for their stability permanent 
social groupings secured by some controlling power. 

On this rests, in large part, the long continuance of 
slavery and serfdom, although it cannot be said that the 
necessity for union of labour originally created these in- 
stitutions. Nevertheless wherever property in man and 
hereditary dependence of the labouring population have 
existed, we notice in the early stages that master and slave 
are distinguished but slightly from one another; that they 
perform their work together; that the dependent class 
is, in numbers, hardly stronger, indeed often weaker, than 
the ruling one. But in the course of time this is changed; 
the enslaved part of the population becomes more numer- 
ous, though less through natural internal increase than 
artificial augmentation from without by means of wars 
of conquest, men-stealing, *the slave trade, and misuse of 
power against weaker freemen. At the same time the 
class of propertied freemen is ever more sharply distin- 
guished from that of the unfree; labour becomes in the 
eyes of the former a disgrace, while for the latter it de- 
velops into a burden of constantly growing oppressive- 
ness. A deep gulf rends society, and there is no means 
of bridging it other than release from the condition 



320 ORGANIZATION OF IVORK AND 

of compulsory labour. Frequently even this does not 
suffice, as is shown, for instance, by the sharp distinction 
between freemen and freed men among the Romans. 

The necessity for this graded progress lies in the tech- 
nical conditions affecting the developed forms of labour 
in common. The natural consequence of the imperfect 
character of the implements * is that larger tasks can be 
accomplished only through the application of combined 
human labour on a large scale. Each advance of the 
household economy thus necessarily presupposes an in- 
crease in the number of its unfree workers. Each rise in 
the standard of life of the ruling class involves a waste of 
human material, which, according to our conceptions, is 
monstrous. To realize an effective union of labour this 
material must be organized and disciplined. 

The necessity of working slaves in gangs has from time 
immemorial been deduced from their unreliability and lazi- 
ness which compelled the strict supervision of their work. 
It is indeed true that these features everywhere character- 
ize servitude. But not it alone; they are rather phenom- 
ena incident to a half-developed culture in general, which 
at such a stage may be found even among free people. 
Moreover the slave-holder applies the system of division 
of labour along with labour in common whenever this 
can result in such an assignment of definite duties to the 
individual workman that he can be made responsible for 
the performance of them.*^ But in the sphere of produc- 
tion the allotment of particular tasks to the individual is 
usually either impossible or inadvisable, because profitless. 

^Comp. also A. Loria, Die Sklavenwirthschaft im modem, Amerika 
u. im europdisch, AUertum, in Ztschr. f. Sozial. u. Wirthschaftsgesch., 
IV, pp. 68 ff. 

*This indeed takes place especially with housework and personal 
services. See above, pp. 98, 99, 299, joa 



THE FORMATION OF SOCIAL CLASSES. S^^ 

Thus at this stage we see labour in common assuming 
extensive proportions and becoming by far the most 
potent organizing principle of unfree labour. 

David Hume long since remarked ® that slavery neces- 
sitated a strict military discipline; and our investigations 
are corroborative of this observation. 

In early Egypt " each of the great administrative offices 
possessed its own craftsmen and workmen. These were 
divided into bands. We even meet with such a company 
on the estates of the more prominent men of the ancient 
empire, and notice how, led by their ensign, they draw 
up on parade before the lord of the estate. The galley- 
slaves of every larger ship likewise form a company, and 
even the demons that nightly propel the ship of the sun 
through the lower world bear this name. The craftsmen 
of the temple and of the necropolis are similarly organized. 
The Egyptian magistrate cannot think of these people of 
lower rank otherwise than collectively; the individual 
workman exists for him no more than the individual sol- 
dier exists for our high army officers. Just as these free 
or half-free workers always appear in companies, so the 
slaves of the temple and the necropolis and the unfree 
peasants of the manors are duly organized in military 
fashion and regarded as a part of the army." ^ 

The large Roman slave estates exhibit like phe- 
nomena. On the rural estates the unfree workers are 
divided into groups according to their occupation; each 
group falls again into trains of not more than ten men 
under a " driver "; the Tnllicus is commander-in-chief over 
all. Their day's work is performed with military discip- 
line; at night they are lodged in barracks. In the 
wealthiest homes the urban family likewise exhibits such 

* Essays, p. 252. 

* Erman, Aegypten und dgyptisches Leben im AHertum, pp. 180-186. 



322 ORGANIZATION OF fVORK AND 

ordered groups; in the Imperial household the separate 
slave groups are expressly designated colleges or cor- 
porations.® 

We see here how the need for labour in common led 
to permanent organizations among the unfree; and this 
need was met in the same way by the agriculturalist of 
later Roman times, by the manorial constitution of the 
Middle Ages, and by the more modem servile tenure. In 
each of these the labourers necessary for the large rural 
estates were united into distinct corporate groups at- 
tached to the soil, in order that they might always be 
ready for seed-time and for harvest. One can really say 
that manorial servitude, attachment to the soil, and per- 
sonal subjection ^ owed their ascendancy to the necessity 
of labour in common, and that their great extension and 
long duration were conditioned by this necessity. 

A reaction of labour in common upon the organization 
of society is thus established beyond doubt, giving it not 
merely a peculiar socio- judicial impress, but also deeply 
influencing the mental disposition of the associated work- 
ers. One of the keenest observers of agrarian conditions 
in North Germany*® found as a prominent trait in the 
character of the peasants " that they cling very closely to 

■Thus mention is made of collegia (corpora) lecticariorum, taber- 
naclariorum, cocorum, pragustatorum, decuriones or propositi cubiculario- 
rum, velariorum, tricUniariorum, siructorum, ministratorum, balneariarum, 
unctorum, etc. On all this comp. Marquardt, Privatleben dcr Romer, pp. 
144 ff., 154. The remarks in text do not contradict what was said above 
on pp. 98, 99 regarding division of labour in the slave family of the 
Romans. This sprang from the necessity of having for each piece of 
work required by the household a responsible person — ^not from the 
khowledge of the greater productivity of divided work — while labour 
in common had its basis in technical considerations. 

• Horigkeit, SchollenpAichtigkeit, Leibeigenschaft. 

" Christian Garve, Ueber d. Charakter d. Bauem u. ihr Verhdknis gegen 
d. Gutsherm u, gegen d, Regierung (Breslau, 1786), pp. 14 ff. 



THE FORMATION OF SOCIAL CLASSES. 323 

each other." They live much more sociably among them- 
selves than the ordinary citizens of the towns. They see 
each other daily at each piece of demesne work, in summer 
in the field, in winter in the bam and in the spinning-room. 
Like soldiers, they constitute a corps, and like them gain 
an esprit de corps. The same may be said of all unfree 
conditions; uniformity and the disciplining of work create 
uniform herd-like masses which become more dull and in- 
dolent the more hopeless their condition. 

This explains the small productiveness of their labour, 
which in turn leads to inhumanly harsh treatment, re- 
ducing the labourers to the level of the animal. Genera- 
tion after generation of like labour perpetuates the same 
way of thinking, the same feelings and sensations towards 
the oppressors. The ruling race is now markedly distinct, 
both intellectually and physically, from the subject one, 
just as the vigorous tree in the forest stands out from the 
weakling. But in this evolutionary process causes and 
consequences are confused as in a tangled skein; one 
perceives only a labyrinth of economic and social factors, 
acting and reacting, and nowhere a thread to guide with 
certainty the investigating eye. There are close relation- 
ships existing between the two spheres; that is all that 
we can with some measure of assurance determine. 

The problem offered by the third primal form of or- 
ganization of labour, division of labour ^ would seem rela- 
tively much easier of solution. Moreover, each indi- 
vidual in the world of to-day has a certain interest in it, 
inasmuch as he is personally affected by it. For every- 
one, if he does not wish to be a useless member of society, 
has to accommodate himself to a particular task; and the 
more completely he succeeds in this, the more diversified 
do men themselves become in their every action and 
thought. 



3*4 ORGANIZATION OF l^ORK AND 

The German census of occupations of 1895 recorded in 
all 10,298 distinct trade designations. Now one may as- 
sume that different names are current for many trades in 
different parts of the country, and that a deduction is ac- 
cordingly to be made for double counts. On the other 
hand, one must also remember that very different kinds 
of work, especially in the public service and the liberal 
professions, are designated by the same name, and that 
the numerous individual tasks which have arisen within 
the separate industrial undertakings through division of 
work and which have been transferred to special work- 
men, can be but imperfectly taken into account in the re- 
turns. Thus the census figures should rather be increased 
than reduced. We have thus in round numbers 10,000 
kinds of human activity, each of which can become in our 
modern society a life-work, and subject the whole person- 
ality to its sway. 

New special trades, moreover, are being formed con- 
tinually." Each new process of production, each advance 

" From 1882 to 1895 the number of trade designations in the Ger- 
man census of occupations has been increased by 41 19. The returns 
were, as follows: 

Accordiaif to the Census 
For the Class of Occupstioas. of Occupations for 

188a. 1895. 

A. Agriculture, gardening, cattle-raising, forestry, 

fisheries 352 465 

B. Mining and quarrying industry, and building 

trades 2,661 5,406 

C. Trade and commerce I»2i5 2,216 

D. Domestic services and wagework of varying kind. 75 82 

E. Military, court, civil, and ecclesiastical service, 

liberal professions 1,876 2,079 

Total 6,179 10,298 

How far this growth in figures is to be traced to an actual increase in 
trades, how far to greater exactness in statistical census work, cannot 
be determined. A part of the difference, however, is certainly to be 
attributed to increasing division of labour. 



THE FORMATION OF SOOAL CLASSES, 325 

of technique and science, is subjected to the universal di- 
vision of labour. Thinking and feeling men are thus 
forced into the restricted field of trade interests of. the 
narrowest and pettiest sort. The time foreseen by Fer- 
guson, when even thinking would become a special busi- 
ness, has long since been reached.^ ^ The scope of uni- 
versal human interests grows narrower the greater the 
divergence of the special interests of the numerous spheres 
in life from one another, and the greater the severity of 
the struggle for existence. 

The differences among men due to nature and culture 
without doubt assist this divergence in the most varied 
spheres of life; yet, in our opinion, this is true to a much 
smaller extent than is frequently assumed. Of course, as 
everyone knows, a jockey must differ from a carrier, a 
brewer from a tailor, a dancer from a singer, a poet from 
a merchant, if he is to be competent for his vocation. But 
what natural talents cause one man to appear destined to 
be an inspector of diseased meat, another a bookbinder, 
and a third a chiropodist, hosiery manufacturer, or tobac- 
conist, will likely be as difficult to fix as to determine 
beforehand the success of a particular individual in any 
given liberal profession. 

Although, then, many classes of occupations are 
adapted to bringa particular talent to the highest develop- 
ment, with many others the presence of such a talent will 
be of no perceptible importance. All, however, through 
continuous practice and use, will produce a certain 
differentiation of the men devoting themselves to them; 

" Most notoriously in politics, where the majority of men procure 
their ideas ready made from some newspaper editorial. But also to no 
inconsiderable degree in scientific circles, where on this account the 
last is always right; for example, the reviewer of a book over the au- 
thor. 



326 ORGANIZATION OF IVORK AND 

certain organs will become enfeebled through lack of use, 
while others, through constant exercise, will be developed 
to greater perfection; according to his task the individual 
will be attuned physically, intellectually, and morally, to 
a definite key; through his occupation he will be given a 
particular impress which will often be even externally dis- 
cernible. This we all recognise when we come into con- 
tact with strangers and involuntarily classify them to our- 
selves according to callings. 

With this personal differentiation, however, the eco- 
nomic graduation is transferred also to society at large. 
Similar occupations and views of life, similar economic 
position and social habit lead to a new distribution of 
social groups. They produce classes based on occupation 
and a community of interests which dominate them even 
in their most minute social ramifications, and are strong 
enough to cover up inherited differences in position 
due to birth, or to reduce them to insignificance. We 
have even seen how these new social aggregations 
reach out beyond the political boundaries, and how the 
social interests and feelings of kinship resting on di- 
vision into trades overtop those of nationality based upon 
similarity of blood. 

Under these circumstances we may raise the question, 
which recent biology has brought into close connection: 
whether, and to what extent, in a society with free choice 
of occupation, the personal variations developed through 
division of labour are hereditary y just as under the system 
of castes and of classes according to birth such peculiar- 
ities are transmissible. In this it is not merely a question 
of natural capacities which may be utilized in one's occu- 
pation and in which the possibility of hereditary transmis- 
sion — though not more — is readily admitted. It is a 
question rather of the whole physical and mental aptitude 



THE FORMATION OF SOCIAL CLASSES. 3^7 

for a vocation, of the skill gained through accommodating 
oneself to a circumscribed task, of the intellectual plane 
consequent upon such work, of the conception of life, and 
the direction of the mind resulting from the character of 
one's vocation. 

From the latter point of view, ever since Shakespeare's 
" Winter's Tale," the problem has frequently been treated 
in literature. Generally this has been done by making 
educational influences that counteract upon the character 
and social position of the parents determine events. 
Views as to the issue have greatly changed in the course 
of the last century. It would certainly be a profitable 
undertaking for a literary historian to take up this prob- 
lem of education and heredity, and investigate more 
closely the dependence of literature upon the spirit of the 
times and upon the position in life of the writers.^* While 
Lindau in Countess Lea makes the daughter of the 
usurer develop, in spite of the paternal education, into a 
paragon of nobleness, in a story by Arsene Houssaye {Les 
trois Duchesses), of three children interchanged directly 
after birth, the son of the peasant woman remains peasant 
in understanding and in way of thinking, although edu- 
cated as a prince; the daughter of the frivolous actress 
becomes a courtesan, and the daughter of the duchess, 
even in humble surroundings, displays the native eleva- 
tion of her character. 

The question has also been touched upon in numerous 
ways in more serious literature. But a short time ago 
W. H. Riehl, in his Culturgeschichtliche Characterkopfe, 
drew a contrast between the "peasant youngsters with 
limited capabilities " who had graduated from the gym- 

''The latest treatment of this subject is to be found in Ludwig 
Ganghofer's tale, Der Klosterjdger (Stuttgart, 1893). It is exception* 
ally healthy and subtle. 



32^ ORGANIZATION OF fVORK AND 

nasium with highest standing and the "intellectually 
highly trained sons of cultured parents," between whom, 
class for class, there arises an insurmountable wall. The 
former, he believes, would develop at the university into 
mediocre students, whom the " cultured son of cultured 
parents," if he went to the university at all, would soon 
overtake. Finally the former peasant youth becomes " a 
very mediocre though clerically efficient civil servant." 
What becomes of the son of the cultured parents, " who 
has already been favoured by the manifold educational in- 
terests of his parents' home," we are, unhappily, not in- 
formed. 

The first to discuss the subject with a claim of strict 
scientific treatment,^* which, to be sure, is not made in 
the above case, was Professor Gustav Schmoller, who, in 
a very confident manner, rendered his decision that " the 
adaptation of individuals to various activities, increased 
through heredity during centuries and thousands of years, 
has produced men of ever more individual and diverse 
types." All higher social organizations, it is claimed, 
rest upon continued differentiation produced by division 
of labour. "The castes, the aristocracies of priests, of 
warriors, of traders, the guild system, the whole constitu- 
tion of labour to-day are but forms differing according to 
the times, which division of labour and differentiation 



14 



Schmoller has objected to this expression in his review of my 
book in Jhrb. f. Gesetzg. Verw. und Volksw., XVII (1893), pp. 303 ff- 
He desires to have his remarks regarded as but "a kind of essay in 
philosophical history." I can perceive in this characterization no re- 
pugnancy to the expression used by myself. Nor can I discover that 
the further remarks of Schmoller in the paper cited have furnished 
proof that I have misunderstood him in essential points. I believe, 
therefore, that I am acting most correctly in allowing the following 
remarks to appear again word for word as they stood in the first edi- 
tion, and in directing the attention of the reader to Schmoller's re- 
marks on the same in the iTrticle indicated. 



THE FORMATION OF SOCIAL CLASSES. 3^9 

have imprinted upon society; and each individual has ar- 
rived at his peculiar function not merely through indi- 
vidual adroitness and fate, but also through his physical 
and mental disposition, his nerves, and his muscles, which 
rest upon hereditary tendencies and are determined by a 
causal chain of many generations. The differences in social 
rank and property, in social esteem, and in income are only a 
secondary consequence of social differentiation,^* ^^ 

One will perhaps expect that the proof for these sur- 
prising sentences has been attempted with the help of 
biology. But, aside from cursory reference to biological 
analogies, that path is avoided. Yet it would certainly 
have been useful to pursue it further, because it must 
have led inevitably to a point where the conception of 
heredity must needs Have been defined and its sphere 
marked oflF from that of imitation and education.^ ^ 

On this account we also will have to avoid this path, 
and enter upon an examination of the elaborate historical 
and ethnographical material that Schmoller adduces for 
his assertions. 

Such historical proofs are of a nature peculiar to them- 
selves. To the eye of one gazing backwards things get 
shifted from their proper place. Cause and effect appear 
equally near in point of time. One finds oneself in a posi- 

" Comp. Schmoller's articles on the division of labour in his Jhrb., 
XIII, pp. 1003-1074; XIV, pp. 45-105; and a short summary of his 
conclusions in the Preuss. Jhrb., LXIX, p. 464. [See further his 
Grundriss, pp. 395-411. — Ed.]. 

^*Such an attempt, though indeed with but meagre results, is to be 
found in Felix, Entwickelungsgesch, d, Eigentkums, I, pp. 130 S. Among 
the more recent biologists this point in the problem of heredity is 
really no longer a matter of controversy; especially Weismann (The 
Germ-plasm, Eng. ed., London, 1893) has decidedly contested the trans- 
missibility of acquired characteristics. Comp. also Galton, A Theory 
of Heredityy in Journal of Anthropolog. Institute, V, pp. 329 flf. ; James, 
The Principles of Psychology, II, 678. 



330 ORGANIZyfTION OF IVORK AND 

tion similar to that of the man who looks away into the 
distance and sees a church steeple that really rises far 
behind a group of houses apparently standing directly 
over the nearest building. 

After a similar fashion, we fear, SchmoUer in the critical 
instances of his comprehensive investigations has viewed 
the causal relationship of the historical processes in an in- 
verted succession as regards reality. So far as these are 
occurrences that do not reach back into epochs be- 
yond the range of historical investigation, such as the 
origin of castes, of the priesthood, of the oldest nobility, 
we would venture to believe that one might unhesitatingly 
reverse his surprising conclusion and say: the diversity of 
possession and of income is not the result of division of 
labour, but its chief cause. 

For the past, in so far as it lies open to our eyes, this 
can be demonstrated with absolute certainty. Inequality in 
the extent and tenure of landed property forms among the 
ancient Greeks and Romans, and even among our own 
people from the early Middle Ages onward, the basis of 
class organization. The noble, the peasant class, the class 
of villeins and serfs are at first mere classes based on prop- 
erty; it is only after a considerable time that they de- 
velop into a species of classes based on occupation.^ ^ 
When in the Middle Ages along with the rise of the crafts- 
man class the definite formation of trades sets in, it pro- 
ceeds again from distribution of property. The demesne 
servants, the landless villeins who have learned an indus- 
trial art, begin to turn their industrial skill to independent 
account. The industrial process followed must adapt itself 
to their poverty; it is pure wage-work, in which the work- 

"The presence of the unpropertied noble in the service of others 
ipiinstadet) is a proof, not against, but for, this conception. It 
would be inconceivable that the landed noble had not preceded him. 



THE FORMATION OF SOUAL CLASSES. 331 

man receives the raw material from the customer. Only 
later do we have a real division of production between 
agriculturalist and craftsman. The latter acquires a busi- 
ness capital of his own. But how trifling this is, is best 
indicated by the circumstance that, as a rule, the crafts- 
man works only on ordered piece-work, and that the 
whole industrial process for transforming the raw into the 
finished product lies usually in 0}ie hand.^® The industrial 
undertakings were exclusively small undertakings. Where 
the great extent of the sphere of production of a handi- 
craft called for an increased supply of capital, men did not 
turn to production on a large scale with subdivision of 
work, but to specialization which limited the demand for 
capital and kept the business small. 

As one observes, each step taken by mediaeval division 
of labour in industry was conditioned by the possession of 
wealth. It is the same^with trade. The trading class of 
the Middle Ages is derived from the class of urban land- 
owners, who had become, through the introduction of 
rents on houses and the practice of rent-purchases, pos- 
sessors of movable capital. It is from this class of stock- 
holders and tradesmen that the present manufacturing 
class has sprung since the seventeenth century. Through 

^The longer the duration of the process of production the smaller 
the business capital that the single producer requires, but the greater 
the mass of labour which the completed product contains. In the 
Middle Ages, to cite a very familiar example, the shoemaker was fre- 
quently tanner as well. The whole process of industrial elaboration 
from the raw hide to the finished footwear thus lay in one hand. As- 
suming new that the tanning of the hide required half the time that 
was necessary to its transformation into shoeware, a shoemaker de- 
siring to carry on tanning alone would have required thfee times as 
-nuch business capital as the tanner who at the same time made shoes. 
But if he wished merely to make up into shoes leather already tanned, 
his business capital must amount to one and a half times the former, 
together with wages and profits. 



332 ORGANIZATION OF IVORK AND 

the fertilizing of industry with their capital, the two new 
forms of division of labour — ^subdivision and displacement 
of labour — arise, and the division of production for the 
first time realizes its full efficiency. Half-manufactured 
products now wander in masses from workshop to 
workshop; in each place they become capital, in each 
they yield a return; from one department of pro- 
duction to another fresh outlays in interest and other 
charges are added, and through them profits on cap«- 
tal are made.*® Subdivision of labour presupposes a 
class of non-propertied wage-workers. This class comes 
from that section of the craftsmen who, through the capi- 
talistic character assumed by division of labour, have be- 
come incapable of competing, and from the landless 
peasant population. 

In industry, indeed, the dependence of division of labour 
upon possession of property becomes especially manifest. 
In the Middle Ages each advance of industrial division of 
labour augmented the number of urban "livelihoods," 
because it diminished the business capital; at the present 
time the progress of division of labour diminishes the num- 
ber of independent existences since it increases either the 
fixed or the business capital, or both. In the Middle Ages 
the effort was made to keep each industrial product as 
long as possible in one establishment in order to embody 
in it as much labour as was feasible; nowadays, by di- 
vision of work, the business capital is carried with the 
utmost rapidity through the separate stages of production 
in order to make the relation between interest expended 
and profit realized the most favourable possible. In the 

"The connection of capital with division of labour has been pre- 
sented in a masterly manner by Rodbertus {A%is d. litter. Nachlass, II, 
pp. 255 fl.); but in this he has not adequately distinguished the dif- 
ferent kinds of division of labour. 



THE FORMATION OF SOCIAL CLASSES. 333 

Middle Ages dearth of capital led to specialization; in our 
time abundance of capital impels to subdivision of work 
and displacement of labour. 

Thus from the varied distribution of property have the 
general features of our organization of society according 
to occupation been developed historically; and on this 
foundation, which our present industrial organization is 
ever strengthening and solidifying, they continue to rest. 
The latter is explained very simply from the following cir- 
cumstance: I. Every vocation under our industrial or- 
ganization yields an income; and only the propertied per- 
son is in a situation to seek out for himself the more lucra- 
tive positions within the universal organization of labour, 
while the unpropertied person must be content with the 
inferior positions.^® 2. Property itself, by virtue of its cap- 
italistic nature, furnishes an income to its owner, even 
without work on his part, and transmits itself from gen- 
eration to generation with this capability. In so far as 
our propertied classes are also social classes according 
to occupation, they are not such because their occupation 
creates property, but rather because property determines 
the selection of a vocation, and because as a rule the in- 
come that the calling yields is graded much the same way 
as the property on which the vocation is foun4ed. 

True, there is no novelty in this statement. Each 
of us acts conformably to this view. Daily experience 
readily suggests it; and scientific political economy 
has always recognised it. The whole wage-theory itself 
rests on the assumption that the son of the workman can 
become nothing else than a workman. This is a conse- 

*^This means, then, "that those whom poverty drives to seek a 
profitable vocation are compelled by their very poverty to abandon 
that vocation." Lotmar, Die Freiheit d. Berufswahl (Leipzig, 1898), 
p. 27. 



334 ORGANIZj4TIOS OF IVORK AND 

quence of his poverty, not of hereditary adaptation to his 
trade. Must one then really prove now for the first 
time that occupations whose inception and conduct re- 
quire capital, or whose acquisition demands large outlays, 
are as good as closed to those without capital? The 
much-boasted " freedom of enterprise " thus exists only 
within very narrow limits. In very exceptional cases 
these indeed are now and then transgressed; but as a rule 
it is not the particular vocation, but rather the general 
vocational class ^^ to which the individual is to belong in 
society that is indicated for each person by the wealth of 
the paternal house. The " social rank " that in popular 
estimation is enjoyed by a particular class, however, can 
hardly be maintained without corresponding financial 
equipment — a proof that it also is not a secondary conse- 
quence of social differentiation (resting upon division of 
labour), but essentially a child of the rational union of 
wealth and vocation. 

No matter how many vocational classes may be dis- 
tinguished in society, occupations of very diverse charac* 
ter will still be represented in each, and between these 
callings a continuous exchange of labour will take place. 
This exchange extends as far as the classes of work de- 
mand approximately the same equipment of wealth, and 
as far, therefore, as they stand in the same " social rank "; 
one might also say that it extends as far as people marry 
among each other, or regularly associate with one an- 
other, or as there is approximately the same plane of cul- 
ture. All these things stand together in a mutual rela- 
tionship. It is an every-day occurrence for a high public 

*^ On this concept, in which we attempted to express the reciprocally 
conditioned existence of property and vocation long before we were 
acquainted with Schmoller's work, compare my Bevolkerung d. Kantans 
Basel'Stadt, p. 70. 



THE FORMATION OF SOCIAL CLASSES. 335 

official to destine his son for agriculture in order, later on, 
to purchase him an estate, for the son of a large land- 
holder or manufacturer to enter upon an academic career, 
for the son of a clergyman to become a civil engineer, 
the son of the engineer a physician, the son of the physi- 
cian a merchant, the son of the merchant a lawyer or an 
architect. Just as easy and frequent is the transition 
from peasant to schoolmaster or to brewer, from baker to 
watchmaker, from blacksmith to bookbinder, from miner 
to factory-hand, from farm-hand to station-hand or coach- 
man, etc. We all look upon these transitions, in spite of 
the gfreat differences in labour skill, as socially proper and 
industrially unobjectionable, although there can hardly be 
men " differentiated " more widely through division of 
labour than a statesman and a farmer, a manufacturer and 
a professor, a merchant and an architect, and so forth. 
When the son of the manufacturer in turn becomes 
manufacturer, and the son of the peasant again a peasant, 
we know that in many cases the financial means once 
consonant with this vocation have dictated the occupation 
without regard to the fitness or unfitness of the individual 
for the role thrust upon him. 

This glance at practical life must restrain us from con- 
ceiving in too narrow a sense SchmoUer's theory of the 
hereditary transmission of personal differentiation conse-* 
quent upon division of labour. That the son of the shoe- 
maker by virtue of inherited adaptation should be in a 
better position to produce shoes than, let us say, picture- 
frames; that the clergyman's son, though his father had 
been taken from him on the day of his birth, will, of all 
classes of occupation, exhibit the greater natural aptitude 
for the clerical calling, cannot possibly be meant by that 
theory, even if in the last-mentioned case the forefathers 
of the clergyman during the previous two centuries had 



336 ORGANIZATION OF IVORK AND 

handed down the spiritual office to each other from gen- 
eration to generation. For if we hold strictly to the bi- 
ological idea the adaptation to occupation would neces- 
sarily increase from age to age, and reveal itself in 
continually improving performance of duties. It will, 
however, hardly be seriously maintained that the numer- 
ous clerical families of Protestant Germany, who are in 
the position just described, furnish to-day relatively better 
pulpit speakers and more efficient pastors than in the 
seventeenth century. 

In the domain of the guild handicraft of our towns, in 
consequence of the jealous exclusiveness of the different 
trades, the positions of master-craftsmen, with but few ex- 
ceptions, have been actually passed down from father to 
son from the sixteenth to the eighteenth century. The 
technique, however, not only has not improved, but has 
lamentably degenerated, and now languishes, as Schmol- 
ler himself in an earlier treatise has shown.^^ Far from 
augmenting the technical acquisitions of their fathers, the 
sons have not even been able to maintain the standard of 
professional aptitude reached by them. 

We must therefore look upon the new theory, if we 
would not be unjust to it, as referring to the inheritance 
of bodily and intellecttuil characteristics by tlte members of 
social classes grouped according to occupatiofu But these 
classes are^ as a rule, likewise based upon property and 
income, since the standard of their life, both material and 
intellectual, is conditioned by property and income. Ac- 
cordingly one must demand of the originator of the the- 
ory to distinguish between the consequence of the charac- 
ter of sustenance and education rendered possible for each 
class by the possession of wealth, and the result of heredi- 

" Zur Gesch, d. detUsch, Kleingewerbe im /p. Jhdt,, pp. 14, 667 ff. 



THE FORMATION OF SOCIAL CLASSES. 337 

tary adaptation to occupation. If such a distinction of 
the probable and possible causes is not undertaken, or if 
without examination there is ascribed to division of labour 
that which can be traced back with greater probability to 
the apportionment of wealth, the whole theory must be 
content in its undeniable weakness in " historical proof " 
to be treated as an inexact Darwinian analogy, as a thesis 
advanced without proof. 

That within a whole social class of this kind a trans- 
mission of the "bodily and intellectual constitution," of 
the " nerves and muscles " takes place from one genera- 
tion to another no one has as yet doubted. One may in- 
deed term this heredity, but in this he must* not overlook 
that each fresh generation must be raised through theo- 
retical and practical education to the intellectual and 
moral level of the parents. Though in this the elements 
of culture " fly to " them, to use Riehl's expressive 
phrase,^* though the example of their surroundings in- 
cites them to imitation, though much is appropriated 
without trouble which under other conditions must first be 
learned with effort, it is still a question of the acquired, 
. not of the innate. This holds to a certain extent even of 
the bodily constitution — ^the " nerves and muscles " — so 
far as it rests upon the character of sustenance and edu- 
cation.^* 

Elements of adaptation to a vocation can certainly be 
transferred by the indicated paths of " unconscious ab- 

*• AfiAiegen. 

•• Schaffle, Ban u. Lehhi d, soz, K'drpers (i. Aufl.), II, p. 201, desig- 
nates that the physical side of pedagogy. He says: "The physical 
education of each new generation and its schooling in the external 
graces of the parents or ancestors comes as an immense additional task 
to the procreative activity of the sexes. ... In this second act 
physical adaptations are obtained that were unknown to the parents 
themselves." 



33^ ORGANIZATION OF IVORK AND 

sorption " and imitation, just as well as other elements 
of education. But this process is fundamentally different 
from inheritance in the biological sense.^' That which in 
this sense is said to be hereditary must make its appear- 
ance even when the offspring are completely removed at 
the moment of birth from the influence of their pro- 
genitors. 

We know not whether there are people who consider 
the physical and intellectual peculiarities constituting 
the plane of culture of our six or eight vocational classes 
in society as transmissable in the sense that they must ap- 
pear among the descendants of each class even when 
brought up within another class. It is only individual in- 
stances of this kind that practical life is ever presenting; 
and as yet no one has taken the trouble to collect them. 
They are generally cases of children of the humbler classes 
who are brought up or formally adopted by members of a 
higher class. There will scarcely be anyone bold enough 
to maintain that these persons, artificially united to social 
groups of higher rank, are later on distinguishable from 
the members of these groups by birth by reason of less 
business ability or of a lower plane of culture. 

A further series of observations of this nature is offered 
by the instances in which descendants of one class have by 
their own energy raised themselves into a higher class. 
Everyone knows what difficulties the era of capitalistic 
production opposes to such an attempt, and frequently 
only too successfully. Everyone, too, can readily call up 
the picture of the " upstart " who, with all the technical 

**This latter is the real question with Schmoller, as he plainly indi- 
cates in Preuss. Jhrb., Vol. 69, p. 464. The sociological conception of 
inheritance which Schaffle has constructed in works cited (II, pp. 
208 ff.) is not treated by Schmoller, though many of his remarks re- 
call it 



THE FORMATION OF SOCIAL CLASSES. 339 

ability he shows for his trade, is defeated in his effort to 
reach the intellectual and moral level of his new class. 
This serves again to illustrate the truth that the adapta- 
tion to an occupation enjoined by division of labour — the 
prime condition of business success — is accomplished by 
each individually, and without too much difficulty. But 
the moral and intellectual adaptation demanded by the 
plane of culture of the class ripens slowly even amid 
favourable surroundings, and comes to full maturity only 
in the second or third generation. 

A strict proof of the fallacy of SchmoUer's theory 
of heredity cannot be adduced; but the proofs hitherto 
advanced in favour of its accuracy fall equally short of 
conclusiveness. Before venturing to dogmatize one 
would perhaps have to pass in review the great men of a 
nation and note the vocations of their parents, and the 
number who have issued from classes of humble occupa- 
tion. At the same time one would need to determine for 
the different classes the degree of probability of their 
members attaining a prominent position in which they 
alone could display high ability. Finally one would 
have to ascertain what relation the number of promi- 
nent men who have actually come forth from any given 
class of tradesmen bears to the number obtained by the 
calculation of probabilities. It does not need to be demon- 
strated that for such an investigation all the data are lack- 
ing. 

But it may be maintained that the new theory contra- 
dicts the belief of modern civilized people based, as it is, 
upon the observation of many generations. 

How often the complaint is made that so much talent 
pines under the weight of adversity? If to this dictum 
we oppose the other that real talent will always find a 
way, we may indeed offer a formula to flatter the egoism 



340 ORGANIZATION OF IVORK AND 

of successful competitors, though in reality it meets all 
too rarely with confirmation. 

Our whole socio-juridical development since the French 
Revolution is based on the assumption that admission to 
every free calling and to all public offices, which latter, 
after all, we still regard as the pinnacle of class divisions, 
shall be free to all. This principle of free choice of voca- 
tion, whose recognition has been gained only after severe 
struggles, would be a great mistake, and every endeavour 
towards its realization lost labour, if beside the inequality 
in distribution of wealth the hereditability of vocational 
aptitudes likewise stood in the way of its establishment. 

Even many of our oldest academic arrangements must, 
in the light of this theory, necessarily appear funda- 
mentally erroneous. To what a high degree the costli- 
ness of preparation narrows admission to the favoured 
positions of the business world is well known. From time 
immemorial, however, a great peril to the efficiency of 
the official and the scholastic class has likewise been per- 
ceived; and an effort has been made to obviate this danger 
through scholarships, free board, remission of fees, and 
similar arrangements for rendering study possible to those 
without means. The practical results of these arrange- 
ments may be a subject of dispute. Yet in judging them 
it is essential to remember that advancement in the voca- 
tion enjoying popular esteem depends not only upon per- 
sonal integrity, but also upon the social education of the 
individual, upon his ability to make his own strength felt; 
that in this imperfect world even the capable man who too 
modestly holds back may all too easily be outdistanced by 
the mediocre man who is boldly self-assertive; that he 
who seeks to climb the social ladder from the lower rungfs 
will find it much more difficult to reach the top than the 
man who starts halfway up. The German language has an 



THE FORMATION OF SOUAL CLASSES. 341 

expression for denoting distinction in a line of business 
which happily characterizes the importance of the personal 
element in the achievement of success. It is sich hervorthun 
[literally, to do oneself forward]. Thus it may indeed be 
that the student sons of the peasant in Riehl's story failed 
to distinguish themselves in their vocations because they 
lacked capacity. It is none the less true that many of them 
missed success because they did not know how to " do 
themselves forward " in the right place, how to bring their 
personality into play. 

In every social grouping in which the occupation exerts 
an influence there is generally formed within the different 
classes a community of feeling that turns instinctively 
against the intruder, and in spite of all his talent frequently 
dooms him to failure; while, on the other hand, it supports 
and carries along weaklings belonging by birth to the 
group in question. Thus, as concerns advancement in the 
public service, which still bears in a preeminent degree the 
sign manual of a class characterized purely by its voca- 
tion, personal and family connections often play, along 
with financial standing, a decisive part. Where these 
become a cloak for nepotism they can indeed impress upon 
it the characteristics of an hereditary class. In the broad 
realm of labour, organized according to occupation and 
extending beyond it, property will indeed remain, as long 
as the present economic system lasts, the prime cause of 
social class-formation. And just such an accessory im- 
portance as fell in the stages of unfree labour to com- 
munity of labour, will here attach to division of labour. If 
the employment is inherited it is not because the adapta- 
tion to the vocation has been inherited, but because the 
property is transmitted by which membership in it is con- 
ditioned. 

The above theory of heredity consequently bears^ 



34* ORGANIZATION OF fVORK AND 

though certainly unknown to its originator, the cheeriess 
lineaments of a social philosophy of beati possidentes. It 
calls to the man of humble birth who thinks he has in him 
the power to occupy a higher position in life: " Abandon 
all hope; your physical and intellectual constitution, your 
nerves, your muscles, the causal chain of many genera- 
tions, hold you fast to the ground. For centuries your 
ancestors have been seris; your father and g^ndfather 
were day-labourers, and you are destined for a like posi- 
tion," We need not recite how the consequences of this 
new theory do violence to our moral consciousness, to 
our ideal of social justice. 

In the state of improved thesis in which it at present 
stands, the theory, in our opinion, falls to the ground from 
the very fact that, as is frequently enough observed, in a 
single generation the whole road from zero to the highest 
point of modem culture, from the lowest to the highest 
stage of division of labour, from the foot to the summit of 
the social ladder is traversed, and vice versa. One must 
indeed wonder that such a theory could originate among 
a people who count among their intellectual heroes a 
Luther the son of a miner, a Kant the son of a saddler, a 
Fichte the son of a poor village weaver, a Winckelmann 
the son of a cobbler, a Gauss the son of a gardener, not to 
mention many others.^* 

There is an old anecdote of a cardinal whose father had 

** Valerius Maximus wrote a chapter (III, 4), de humiU loco noHs^ 
^ clari evaserunt, that begins thus: "Saepe evenit ut et humili loco 
nati ad summam dignitatem consurgant et generosissimarum imaginum 
ioetus in aliquod revoluti dedecus acceptam a majoribus lucem in 
nenebras convertant." — In the most recent presentation of his theory, 
^hich shows considerable modification, (^Grundriss, pp. 596 £F.,) 
Schmoller rests the fact "that talents and great men come from all 
classes of a generally highly cultured society " upon " the peculiar m- 
•Huinces of variaHon.*' But this explains nothing. 



THE FORMATION OF SOCIAL CLASSES. 34S 

tended swine, and a French ambassador filled with the 
pride of noble birth. In a difficult negotiation in which 
the cardinal represented the interests of the church with. 
adroitness and tenacity the ambassador was so carried 
away that he taunted the other with his origin. The 
cardinal answered: " It is true that my father tended 
swine ; but if your father had done so, you would be tend- 
ing them too." 

This little story has perhaps expressed better than a. 
long disquisition could have done what the observation 
of many generations has established: that the virtues by 
which the fathers rise are as a rule not handed down to- 
grandson and great grandson; that even if the occu- 
pation is inherited, the ability to carry it on disappears. 
Each aristocracy, be it aristocracy of property or of occu- 
pation, degenerates in the course of time like the plant 
growing in too fertile- soil. In this it is not at all necessary 
to think of a moral decay; it suffices if the physical and in- 
tellectual powers decline, and procreation grows weak. 
The introduction of uncomipted blood, ascending from the 
lower to the higher vocations, appears then a condition 
fundamental to the healthy exchange of social material. 
The great problem of the century, indeed, we have always 
considered to be the ensuring that a gradual rise in the so- 
cial scale is made possible ; that a continuous regeneration 
of the higher vocational classes takes place. In the caste 
system, which would be a necessary consequence of the 
theory of heredity, we have ever seen the beginning, not 
the end of the progress of civilization. 

We will not allow ourselves to be led astray in this con- 
ception. The solution of the problem just mentioned is 
for modem civilized peoples a question of their very ex- 
istence. For if history has taught anything with insistence 
it is, that for a people that can no longer be renewed 



344 IVORK AND SOCIAL CLASSES. 

from the fresh spring of pure physical and intellectual 
strength flowing in the lower classes, the statement once 
made by B. G. Niebuhr with regard to England and Hol- 
land holds good: the marrow has departed from their 
bones, they are doomed to inevitable decay. 



CHAPTER X. 

INTERNAL MIGRATIONS OF POPULATION AND THE GROWTH 
OF TOWNS CONSIDERED HISTORICALLY. 

All prehistoric investigation, as far as it relates to the 
phenomena of the animate world, necessarily rests upon 
the hypothesis of migration. The distribution of plants, 
of the lower animals and of men over the surface of the 
earth; the relationships existing between the different 
languages, religious conceptions, myths and legends, cus- 
toms and social institutions; all these seem in this one 
assumption to find their common explanation. 

In the history of mankind we have, to be sure, aban- 
doned the view that nomad life is to be regarded as a 
universal phase in the growth of civilization, which each 
people necessarily traversed before making fixed settle- 
ments, and which served, along with the taming of 
domestic animals, as the " natural " pathway of a people 
passing from the hunting stage to agriculture. Ethno- 
graphic research has made it sufficiently clear that all 
primitive peoples, whatever the economic foundations 
of their existence, readily, often, indeed, for very insig- 
nificant reasons, shift their habitations, and that they 
exhibit an extraordinary number of stages intermediate 
between nomadic and settled life.* The northern and 
southern limits of the inhabited world are still peopled 

^ Comp. Z. Dimitroff, Die Geringschatsung d, menschlichin Lebens u. 
ikre Ursachen bet d. Naturvolk. (Leipzig, 1891), pp. 33 ff. 

345 



34^ INTERNAL MIGRATIONS OF POPULATION AND THE 

» 

by races without fixed abode; and even in its midst there 
are broad areas in which a condition of continual migra- 
tion prevails. Most civilized peoples have proverbs or 
other historic bequests from such a time. 

In the German language this far-distant period of univer- 
sal mobility has left distinct traces. The word for " healthy" 
(gesund) meant originally " ready for the road." ^ Gesinde, 
signifying to-day household servants, is, in the olden 
speech, a travelling retinue; companions (Gefahrte and 
Gefahrtin) means, in the strictly literal sense, the fellow- 
traveller. Erfahrung (experience) is what one has ob- 
tained on the journey (fahren); and bewandert (skilled) is 
applied to the person who has wandered much. With 
these the list of such expressions is far from ex- 
hausted. In the general significance attached to them 
to-day the universality of the concrete range of concep- 
tions and observations from which they originally sprang 
finds expression. 

It is natural to suppose that this condition of general 
nomadic wandering, with its deep-rooted nomadic cus- 
toms, did not suddenly cease; that, in all probability, 
the whole course of further development down to our own 
day has been a gradual progress towards a settled condi- 
tion and an ever-closer attachment of the man to the spot 
where he was born. 

Various indications support this view. Among our 
forefathers the house is reckoned movable property; 
and it is demonstrable that many settlements have within 
historic times changed their locations. Despite the lack 
of artificial roads and comfortable means of transporta- 
tion, the individual appears in the Middle Ages much more 
migratory than at a later time. This is supported by the 



' [From senden, meaning to go, to travel. — Ed.] 



GROIVTH OF TOIVNS CONSIDERED HISTORICj4LLY. 347 

numerous pilgrimages that extended as far as St. lago, in 
Spain, by the crusades, by the great bands of travellers, 
the migratory life of king and court, the rights of hospi- 
tality of the marquisates and the developed system of 
escorts. 

Each fresh advance in culture commences, so to speak> 
with a new period of wandering. The most primitive agri- 
culture is nomadic, with a yearly abandonment of the cul- 
tivated area; the earliest trade is migratory trade; the 
first industries that free themselves from the household 
husbandry and become the special occupations of separate 
individuals are carried on itinerantly. The great founders 
of religion, the earliest poets and philosophers, the musi- 
cians and actors of past epochs are all great wanderers. 
Even to-day, do not the inventor, the preacher of a 
new doctrine, and the virtuoso travel from place to place 
in search of adherents and admirers — notwithstanding the 
immense recent development in the means of communi- 
cating information? 

As civilization grows older, settlement becomes more 
permanent. The Greek was more settled than the Phoe- 
nician, the Roman than the Greek, because one was al- 
ways the inheritor of the culture of the other. Conditions 
have not changed. The German is more migratory than 
the Latin, the Slav than the German. The Frenchman 
cleaves to his native soil ; the Russian leaves it with a light 
heart to seek in other parts of his broad Fatherland more 
favourable conditions of living. Even the factory work- 
man is but a periodically wandering peasant. 

To all that can be adduced from experience in support 
of the statement that in the course of history mankind 
has been ever growing more settled, there comes a gen- 
eral consideration of a twofold nature. In the first place 
the extent of fixed capital grows with advancing culture; 



34^ INTERNAL MIGRATIONS OF POPULATION AND THE 

the producer becomes stationary with his means of pro- 
duction. The itinerant smith of the southern Slav coun- 
tries and the Westphalian ironworks, the pack-horses of 
the Middle Ages and the great warehouses of our cities, 
the Thespian carts and the resident theatre mark the 
starting and the terminal points of this evolution. In the 
second place the modern machinery of transportation 
has in a far higher degree facilitated the transport of 
goods than of persons. The distribution of labour de- 
termined by locality thereby attains greater importance 
than the natural distribution of the means of production; 
the latter in many cases draws the former after it, 
where previously the reverse occurred. 

To these statements there are, of course, some consider- 
ations and facts opposed. First, the extent to which man 
was by law tied to the soil in the earlier agrarian period — 
the unfree nature of all his economic and legal relation- 
ships in contrast with the modem freedom of person and 
property. Further, and in part as a result of this, 
we have in modem times the entire dependence of many 
individuals upon movable capital or personal skill. Still 
further, the growing ease of transfer of landed property 
which to-day allows the peasant to convert house and land 
into money and on the other side of the ocean to start 
life anew; while the villein of the Middle Ages could at 
most attach himself as an extra-mural citizen to a neigh- 
bouring town whence he either continued to carry on his 
work in the village personally, or leased it in some form 
or other to a second person for a yearly rent. Finally, 
the increase one observes in the flow of rural popula- 
tion to the towns which has been manifesting itself for 
half a century in a remarkably rapid rise in urban, and 
at some points in a stationary or even declining rural, 
population. With all these circumstances in mind, many 



GR01VTH OF TOIVNS CONSIDERED HISTORICALLY. 349 

consider themselves justified in speaking of the steadily 
advancing mobility of society. 

How are these two series of phenomena to be recon- 
ciled? Is it a question of two principles of development 
mutually opposed? Or is it possible that modern migra- 
tions and those of past centuries are of essentially different 
types? 

One would almost be inclined to believe the latter. The 
migrations occurring at the opening of the history of 
European peoples are migrations of whole tribes, a push- 
ing and pressing of collective units from east to west which 
lasted for centuries. The migrations of the Middle Ages 
ever affect individual classes alone; the knights in the 
crusades, the merchants, the wage craftsmen, the journey- 
men hand-workers, the jugglers and minstrels, the villeins 
seeking protection within the walls of a town. Modern 
migrations, on the contrary, are generally a matter of pri- 
vate concern, the individuals being led by the most varied 
motives. They are almost invariably without organiza- 
tion. The process repeating itself daily a thousand 
times is united only through the one characteristic, that 
it is everywhere a question of change of locality by per- 
sons seeking more favourable conditions of life. 

Yet such a distinction would not be fully in accord with 
the nature of either modem or mediaeval migrations. If 
we would grasp their true importance in historical evolu- 
tion we must first thin out the tangled thicket of con- 
fused contemporary opinions that still surrounds the 
whole subject despite all the efforts of statistics and politi- 
cal economy. 

Among all the phenomena of masses in social life suited 
to statistical treatment there is without doubt scarcely 
one that appears to fall of itself so completely under the 
general law of causality as migrations ; and likewise hardly 



350 INTERNAL MIGRATIONS OF POPULATION AND THE 

one concerning whose real cause such misty conceptions 
prevail. 

Yet, not merely in popular circles and in the press, but 
even in scientific works, migratory instincts are spoken of; 
and thus those movements of men from place to place are 
put without the pale of deliberate action. Indeed, a statis- 
tician once entitled an article in the Journal of the Prus- 
sian Statistical Bureau of 1873 "The Affection for the 
Homestead and the Migratory Instinct of the Prussian 
People," just as if home-keeping depended merely upon 
natural disposition, and the abandoning of it upon an 
irresistible instinctive impulse stronger with one race 
than another! 

In strange contradiction to this, to be sure, is the fact 
that, while the great bulk of official statistical compila- 
tions remains unheeded by wider circles, the publication 
of the emigration returns generally excites a most active 
expression of public opinion. The rising and falling of the 
figures bring fear and hope, approbation and disapproba- 
tion, editorial leaders and speech(js in Parliament. Here 
naturally we hear less about migratory instincts and the 
love of home; people have a vague feeling that behind those 
fluctuating phenomena stand very concrete causes. But 
how little they comprehend the nature of these causes 
is evident when we recall, for example, that a few years 
ago it was a matter of grave debate in the German Reichs- 
tag whether people emigrated because they were getting 
along well or because they were not. 

With regard to this problem one cannot say that as yet 
statistics have succeeded in escaping from the turbid 
waters of confused popular opinions to the firm conclu- 
sions of exact observation. From the statistical stand- 
point migration is above all an economic and social phe- 
nomenon of masses; and statisticians, in our opinion, have 



GROiyTH OF TOiVNS CONSIDERED HISTORICALLY. 351 

been precipitate in abandoning the attempt to discover 
with their peculiar machinery the causes of these mig^- 
tions and turning to investigation by inquiry before the 
resources of the statistical method were exhausted. 

A perusal of the perfunctory remarks that Quetelet* 
devotes to the phenomenon of emigration will readily con- 
vince anyone that his exposition of the subject hardly rises 
above the prosaic commonplace. True, one finds on go- 
ing through the official publications of recent date that 
detailed systematic interrogations on the " causes " and 
*' grounds " of emigration, which would not even perplex 
the less intelligent of the communal officials consulted, are 
by no means infrequent. But one immediately feels that 
such suggestive questions mean the substitution of a series 
of subjective presumptions for the objective results of in- 
vestigation. 

Before resorting, however, to a means of information 
that reads into /the numbers only a strained interpre- 
tation not following of itself, we should rather try 
to determine th;e phenomena of migrations themselves. 
We should classify them according to numerical regu- 
larity, and connect them with other mass-phenomena of 
different times and places accessible to statistics (for ex- 
ample, the density of population, its division into trades, 
the distribution of landed property, the rate of labour 
wages, the oscillation in food prices); that is, undertake 
the statistical experiment of drawing up parallel lines of 
isolated series of figures. 

From even these first steps on the road to an exact 
method we are, however, still far removed. The whole de- 
partment of migrations has never yet undergone system- 
atic statistical observation; exclusive attention has hith- 

•Dk systhni social et des his qui U rigissent, pp. 186-190. 



35* INTERNAL MIGRATIONS OF POPULATION AND THE 

erto been centred upon remarkable individual occur- 
rences of such phenomena. Even a rational classification 
of migrations in accord with the demand of social science 
is at the present moment lacking. 

Such a classification would have to take as its starting- 
point the result of migrations from the point of view of 
population. On this basis they would fall into these 
groups : — 

1. Migrations with continuous change of locality. 

2. " " temporary change of settlement. 

3. " " permanent settlement. 

To the first group belong gypsy life, peddling, the carry- 
ing on of itinerant trades, tramp life; to the second^ the 
wandering of journeymen craftsmen, domestic servants, 
tradesmen seeking the most favourable spots for tempo- 
rary undertakings, officials to whom a definite office is 
for a time entrusted, scholars attending foreign^'fnstitu- 
tions of learning; to the third y migration from place to 
place within the same country or province and to foreign 
parts, especially across the ocean. 

An intermediate stage between the first and second 
group is found in the periodical migrations. To this stage 
belong the migrations of farm labourers at harvest-time, 
of the sugar labourers at the time of the campagne, of the 
masons of Upper Italy and the Ticino district, common 
day-labourers, potters, chimney-sweeps, chestnut-roast- 
ers, etc., which recur at definite seasons. 

In this division the influence of the natural and political 
insulation of the different countries is, it is true, neglected. 
It must not, however, be overlooked that in the era of 
nationalism and protection of national labour political al- 
legiance has a certain importance in connection with the 
objective point of the migrations. It would, therefore, in 



GROIVTH OF TOPyNS CONSIDERED HISTORICALLY. 353 

our opinion, be more just to make another division, 
taking as a basis the politico-geographical extent of the 
migrations. From this point of view migrations would 
fall into internal and foreign. 

Internal migrations are those whose points of departure 
and destination lie within the same national limits; for- 
eign, those extending beyond these. The foreign may 
again be divided into continental and extra-European 
(generally transmaritime) emigration. One can, however, 
in a larger sense designate all migrations that do not leave 
the limits of the G^ntinent as internal, and contrast with 
them real emigration, or transfer of domicile to other parts 
of the globe. 

Of all these manifold kinds of migration, the trans- 
maritime alone has regularly been the subject of official 
statistics; and even it has been but imperfectly treated, as 
every student of this subject knows. The periodic 
emigrations of labour and the peddling trade have occa- 
sionally been also subjected to statistical investigation — 
mostly with the secondary aim of legislative restriction. 
The Government of Italy alone has long been endeavour- 
ing to clear up the subject of the periodic migration of a 
part of her population to other European lands through 
local investigations, exchange of tabulation-cards and 
consular reports. 

The migrations involving permanent and temporary 
transfer of settlement between the different European 
countries are but very imperfectly noticed in the publica- 
tions of the population census by means of the returns of 
births and of nationality. As for internal migrations, they 
have only in rare instances met with serious consideration. 

Yet these migrations from place to place within the 
same country are vastly more numerous and in their con- 



3S4 INTERNAL MIGRATIONS OF POPULATION AND THE 

sequences vastly more important than all other kinds of 
migration put together.* 

Of the total population of the Kingdom of Belgium 
there were, according to the results of the census of 31st 
December, 1880, not less than 32.8 per cent, who were 
bom outside the municipality in which they had their tem- 
porary domicile;* of the population of Austria (1890), 
34.8 per cent. The actual population of Prussia on the 
first of December, 1880, was divided as follows: — 

"«• of Birth. PeraoM. PopdatioB. 

1. In the municipality where enumerated 15,721,588 57.6 

2. Elsewhere in census district (Kreis) 4»599f664 16.9 

3. Elsewhere in enumerated province 4,556,124 16.7 

4. Elsewhere in Prussia 1,658,187 6.1 

5. Elsewhere in Germany 526,037 1.9 

6. In foreign parts under German flag 212,021 0.8 

Of 27,279,111 persons, 11,552,033, or 42.4 per cent., 
were bofn outside the municipality where they were domi- 
ciled,* More than two-fifths of the population had 
changed their municipality at least once. Of the popula- 
tion of Switzerland on the first of December, 1888, there 
were born in the commune where then domiciled 56.4, in 
another commune of the same canton 25.7, in another 
canton 1 1.5, in foreign parts 6.4 per cent J The commune 
in this enumeration is an administrative centre, which in 
many parts of the State embraces several places of resi- 
dence. The figures here given thus exclude altogether 
the numerous class of migrations from locality to locality 
within the commune itself. 

* Comp. now also G. von Mayr, StatisHk u, GeseUschaftslehrt, II, pp. 

116 ff., 354 ff- 
*Annuaire statist, de la Belgique, XVI (1885), p. 76. 
'Ztschr. d. k. preusz. statist. Bureaus, XXI (1881), Beilage I, pp. 

46.47. 

* Statist. Jhrb. d. Schweiz, II (1892), p. 57. 



> 



CROiVTH OF TOIVNS CONSIDERED HISTORJCAUY. 355 

This latter class of internal migrations, as far as we are 
aware, has been but once a subject of investigation. This 
was in connection with the Bavarian birth statistics of 
1 87 1.® According to these the total actual population of 
Bavaria was divided as follows: — 

Place of Birth. No. of Perctpfwhok 

ruux w ovuu Persons. Popnlation. 

1. In the municipality where enumerated 2,975, 146 61.2 

2. Elsewhere in census district (Kreis) I43»i86 3.0 

3. Elsewhere in enumerated province 677,752 13.9 

4. Elsewhere in Bavaria 944,101 19.4 

5. Elsewl^ere in Germany 78,241 1.6 

•6. In foreign parts 44»i50 0.9 

The Bavarian population of 1871 thus appears some- 
what more settled than the Prussian of 1880 and the Swiss 
of 1888, a circumstance perhaps due to the earlier year of 
the census. Nevertheless, two-fifths of the inhabitants 
{1,888,000 out of 4,863,000) were not native to the place 
in which they were living; that is, had migrated thither 
at some time or other. In the larger cities the number of 
people not of local birth amounted to as much as 54.5 per 
cent., in the small rural towns 43.2 per cent.; even in the 
communes of the open country it sank to merely 35.6 per 
cent. 

These are colossal migrations that we are dealing with. 
If one may venture an estimate, the data for which cannot 
be given in detail here, we beHeve ourselves justified in 
maintaining that the number of the inhabitants of Europe 
owing their present place of domicile not to birth, but to 
migration, reaches far over one hundred millions. How 
small do the oft-cited figures of transmaritime emigration 
appear in comparison! ® 

' Die bayerische Bevolkerung nach d. GebiirtigkeU, Bearbeitet von Dr. 

G. Mayr (No. XXXII of Beitrage g, Statistik d. Konigr. Bayem), p. 10. 

• In the seventy years, 1821-1891, the United States of America re- 



35^ INTERNAL MIGRATIONS OF POPULATION AND THE 

That such enormous movements of population must 
draw after them far-reaching consequences is obvious. 
These consequences are chiefly economic and social. 

The economic result of all kinds of migrations is a local 
exchange of labour and, as people cannot be dissociated 
from their economic equipment, a considerable transfer of 
capital as well. Or we may say, since we must presume 
that in these matters also men's actions have definite pur- 
poses behind them, that they bring about more effective 
distribution and combination of labour and capital 
throughout the whole inhabited world. In this regard it 
is indifferent whether labour follows capital or favourable 
natural conditions, or capital seeks unemployed hands. 

Their social result is great shiftings of the population, 
which with an endless, undulatory motion seeks to pre- 
serve the equilibrium between itself and existing advan- 
tages for trade. These shiftings retard the increase in 
population at certain points, and accelerate it at others, — 
at once a thinning out and a concentration. The local 
distribution of the population, which is ordinarily deter- 
mined by natural organic increase, through surplus of 
births over deaths, is broken through. 

But in this very respect there is for the individual State 
an important difference between internal migration and 
emigration. 

The immediate effect of emigration upon the mother 
country shows itself in only one way: it thins out the pop- 
ulation and gives elbow-room to the remainder. That at 
the same time the settlement and development of thinly- 
peopled colonial territories is accelerated only indirectly 
affects the mother country when ultimately by the prac- 
tice of agriculture on a virgin soil the emigrants create a 

ceived from all countries of Europe 13,692,576 immigrants, v. Mayr, as 
above, p. 344. 



GROIVTH OF TOIVNS CONSIDERED HISTORICALLY. 357 

dangerous competition for home agricultural products, or 
by the transference of industrial skill and means of pro- 
duction into foreign lands cut off the market of home in- 
dustry. 

The effects of internal migrations, on the other hand, 
are always of two kinds: those displaying themselves 
at the points of departure; those perceptible at the objec- 
tive points. In the one case they reduce, in the other 
they increase, the density of the population. They 
thus cause, as it were, a division of the popula- 
tion centres and districts into those producing and 
those consuming human beings. Our producing centres 
are generally the country, places and smaller towns; our 
consuming centres, the large cities and industrial districts. 
The latter increase in population beyond the natural rate 
of the birth surplus, while the former remain noticeably 
behind it. Taking a yearly average for the period of eigh- 
teen years from 1867 to 1885, the total population of the 
German Empire has increased by 0.86 per cent, of the 
mean population.*^ Yet when we look at the details we 
see that the average yearly increase amounted: 

In the large cities (pop. 100.000 and over) to a.6 per cent. 

" " medium-sized cities ( " 20,000 to xoo,ooo) " a.4 " " 
•• " small cities ( " 5.000 *' 20,000) " 1.8 " *• 



(• 


" country towns 


( " 


2,000 •• 


5,000) •• I.O " 


*« 


" villages 


(below 


2,000) 


M J .. 



«< 



But of course the phenomenon of inland migrations is 
really not so simple and clear as this row of figures would 
seem to indicate. It certainly vividly illumines the much- 
talked-of " indux to the cities'^ This expression, however, 
tells only half the truth. It overlooks the great number of 
internal migrations that counterbalance one another, and 

^ According to Schumann in Mayr's Allg. statist. Archiv., II (1890), 
p. 518, 



35^ ISTERNAL MIGRATIONS OF POPULATION AND THE 

therefore find no expression in a change in the number 
of inhabitants of individual localities. 

If we take a collective view of the internal migrations 
of a large country, without regard to their effect on the 
distribution of the inhabitants over the surface, their 
routes appear to us as a close variegated web in which the 
interwoven threads cross and recross continually. Into 
the rather simple warp stretched from the country places 
and towns to the large cities and industrial centres is 
woven a many-coloured woof whose threads run hither 
and thither between the smaller centres of population. Or, 
to use a different figure, the broad and majestically surg- 
ing surface-current, which alone we see, is not the only 
one; beneath it numerous lesser currents sport at will. 

Up to the present these latter have received scarcely 
any attention, certainly not so much as they deserve, even 
in cases where they happen to have been statistically as- 
certained. The Bavarian census of 187 1 shows the follow- 
ing situation: 

Reudents native i>^» 
tu Locality of p.^^I^^ Total. 

In the self-governing cities 301,494 361,899 663.393 

In other places of over 2,000 population. 205,887 157,000 362,887 

Total 507*381 518,899 1,026,280 

In the rural municipalities 2,467,765 1,357.981 3*825,746 

Grand total 2,975,146 1,876,880 4,852,026 

From these figures it is plainly evident that the abso- 
lute number of persons who during the last generation 
migrated into rural municipalities is far more than twice 
as great as the number who had migrated to the cities. 
The same relation probably holds good for all larger 
countries. 

But the significant feature in this connection is not that 



CROfVTH OF TOfVNS CONSIDERED HISTORJCALLY, 35S^ 

the country places receive as well as give in this inter- 
change of population; it lies in two other considerations. 
The one is that they give out a larger population than 
they receive; the other, that their additions are made 
chiefly from the rural municipalities, while those leaving 
them find their way in part to the more distant cities. The 
excess of decrease over increase thus accrues to the bene- 
fit of communities of higher order; so much of the popu- 
lation enters into a sphere of life economically and socially 
different. 

If we call the total population born in a given place 
and domiciled anywhere within the borders of the coun- 
try that locality's native population, then according to the 
conditions of interchange of population just presented 
the native population of the country places is greater than 
their actual population, that of the cities, smaller. Thus 
in Bavaria, according to the census of 1871, the native 
population of the rural municipalities amounted to 103.5 
per cent, of the enumerated population, that of the cities 
to only 61 per cent.^^ In the Grand Duchy of Olden- 
burg *^ according to the census of December ist, 1880, 

The influx from other places amounted in the cities to 25,370 persons 
The exodus to '• " " " " cities ** lO.aoS 

The influx from '* ** " •* " country ** 57,366 

The exodus to '• " " •• •* country " 72,528 



*< 
«< 



A balancing of the account of the internal migrations 
thus gives the cities a surplus, and the country municipal- 
ities a deficit, of 15,162 persons. In the economy of popu- 
lation one is the complement of the other, just as in the 
case of two brothers of different temperament, one of 
whom regularly spends what the other has laboriously; 

^ Mayr, as above, pp. 53, 54 of the introduction. 

^ Comp, Statist. Nachrichten uber d. GrossK Oldenburg, XIX, p. 64. 



36o INTERNAL MIGRATIONS OF POPULATION AND THE 

saved. To this extent then we are quite justified from 
the point of view of population in designating the cities 
man-consuming and the country municipalities man-pro- 
ducing social organisms. 

But the total remaining loss of population of the coun- 
try municipalities exceeds the surplus that they furnish 
to the cities, even in the example here given from a small 
State, by almost four times. And the amount that they 
receive from one another is just as great. However large 
this mutual exchange of population by the country places 
,may appear, only a relatively limited scientific interest 
really attaches to it. For here we are dealing with a 
species of migration which arises from the social limita- 
tions of the rural places, and which accordingly gains in 
importance the smaller the communities. In the whole 
Grand Duchy of Oldenburg the number of persons not 
bom where at the time residing amounted in: 



Municipalities of less than 500 inhabitants to 55«o% 

" 500 to 1,000 " " 37.4% 

" " 1,000 to 1,500 " " 41.7% 

" " 1,500 to 2,000 " " 40.4% 

" " 2,000 to 3,000 " " 28.7% 

" 3,000 to 4.000 " " 22.2% 

" 4,000 to 5,000 " " 20.6% 

over 5,000 " " 29.4% 



« « ^..^. m .«u»M » « 



From this we notice that in the smaller municipalities 
(up to 4000 inhabitants), as the absolute size of the mu- 
nicipality increases the influx from other places decreases 
relatively to the native population, while in the larger 
places it increases. 

Mayr has shown that the same holds for Bavaria. There 
in the year 1871 in the larger rural municipalities (of 2000 
and more inhabitants) the number of those resident in the 
place of birth was 66.9 per cent., but in the smaller mu- 



GROHTTH OF TOIVNS CONSIDERED HISTORICALLY. 3^^ 

nicipalities only 64.4 per cent./' while in the cities the 
exact opposite was the case. For in the self-governing 
cities 45.5 per cent, of the population were found to have 
been born where enumerated, but in the other (smaller) 
towns 56.8 per cent. Mayr accordingly sets up the prop- 
osition that in the cities the proportion of persons bom where 
residing decreases with the size of the place, while in the 
rural municipalities , on the contrary, it increases}^ 

There is a very natural explanation for this condition 
of affairs in the country. Where the peasant, on account 
of the small population of his place of residence, is 
much restricted in his local choice of help, adjoining com- 
munities must supplement one another. In like man- 
ner the inhabitants of small places will intermarry more 
frequently than the inhabitants of larger places where 
there is a greater choice among the native population. 
Here we have the occasion for very numerous migrations 
to places not far removed. Such migrations, however, only 
mean a local exchange of socially allied elements. 

This is again clearly shown by the work, already fre- 
quently referred to, on the native-born population of 
Oldenburg. In it the foreign-born population of Wadde- 
warden, Holle and Cappeln, three communities chosen at 

^Die bayer, Bevolkerung nach d. Gehurtigkeit. Introduction, p. 15. 

^^This proposition has been corroborated by the Austrian census of 
1890. According to the excellent treatise on it by H. Rauchberg, Die 
Bevolkerung Oesterreichs auf Grund d. Ergebnisse d. Volkss. v. 31. Des, 
1890 (Vienna, 1895), p. 105, of every 100 persons bom where enumer- 
ated there were in places: 

Of less than 500 inhabitants 65.7% 

" 500 to 2,000 " 73.5% 

" 2,000 to 5,000 " 69.9% 

" 5,000 to 10,000 " 55.6% 

" 10,000 to 20,000 ** 46.4% 



« 



over 20.000 inhabitants 43-i% 



Holle. 
1298 

44S 


Cappeln 

1423 
388 


267 
60.1 


324 
83.5 


178 
39-9 
544 


64 
16.5 

387 


490 
90.0 


33a 
85.9 


54 
10. 


55 

14.1 



362 INTERNAL MIGRATIONS OF POPULATION AND THE 

random, is arranged according to zones of distance from 
the place of birth. The figures are as follows: ** 

Waddewarden. 

Total population 86x 

From other places 270 

Of these Utter there ( j^j^j j^ g 

?oT«/".V^.'r."!iP"«nt ^5.6 

From greater distance. \ ^^^^^ »-^«^ »,^ 

Migrated to other places 400 

Of those up to a dit- ( Absolute number 332 
tance of 9 miles ( Per cent 83.0 

Migrated to a greater ( Absolute number 68 
distance ( Per cent 17.0 

How entirely different are conditions in this regard 
in the capital, Oldenburg, which with its 20,575 inhabi- 
tants is after all to be looked upon as only a small city. 
Of its total foreign-born population (13,364 persons, or 
64.9 per cent.) there come: 

From a Distance of— Persona. Per cent. 

Less than 9 miles 2916 21.8 

From 9-45 •* 5625 42.1 

Over 45 " 4823 36.1 

Here the greater part of the influx of population is from 
a distance; the entry of the stranger-born into a new 
community means at the same time an entry into new 
social and economical conditions; and this urban com- 
munity does not give as many of its native inhabitants 
to other districts as it receives from them.^® On the 
contrary, it absorbs from a wide region round about the 
surplus of emigration over immigration, and repays it 
only in very small part. 

"Statist Nachrichten iiber d. Grossh, Oldenburg, p. 65 [i German mile 
is taken = 4.5 English miles, although actually = 4.6. — Ed]. 

" The city of Oldenburg in the year 1880 received from other munici- 
palities of the Grand Duchy 8,725 inhabitants, and gave up to them only 
1,925. See, as above, p. 212. 



CROfVTH OF TOIVNS CONSIDERED HISTORICALLY. 3^3 

This is the characteristic of modem cities. If in our 
consideration of this problem we pay particular attention 
to this urban characteristic and to a like feature of the 
factory districts — where the conditions as to internal mi- 
grations are almost similar — ^we shall be amply repaid 
by the discovery that in such settlements the result 
of internal shiftings of population receives its clearest 
expression. Here, where the immigrant elements are 
most numerous, there develops between them and the 
native population a social struggle, — ^a struggle for the 
best conditions of earning a livelihood or, if you will, for 
existence, which ends with the adaptation of one part to 
the other, or perhaps with the final subjugation of the 
one by the other. Thus, according to Schliemann,*'' the 
city of Smyrna had in the year 1846 a population of 80,000 
Turks and 8,000 Greeks; in the year 1881, on the con- 
trary, there were 23,000 Turks and 76,000 Greeks. The 
Turkish portion of the population had thus in 35 years 
decreased by 71 per cent., while the Greeks had increased 
nine-fold. 

Not everywhere, to be sure, do those struggles take the 
form of such a general process of displacement; but in 
individual cases it will occur with endless frequency within 
a country that the stronger and better equipped element 
will overcome the weaker and less well equipped. 

In the year 1871, for instance, there were, in round 
numbers, 86,000 Bavarians living in Munich not born 
in the city; and at the same time some 18,000 na- 
tives of Munich were to be found in other places in 
Bavaria. In the year 1890 55.3 per cent, of the popula- 
tion of the twenty-six largest cities of Germany was found 
to have been born in other places, while 22.3 per cent, of 

" Reise in d. Troas im Mai 1881, pp. 29 ff. 



3^4 INTERNAL MIGRATIONS OF POPULATION AND THE 

their native population was living in other parts of the 
empire.*® Still more striking is the fact shown by the 
English census of 1881, that there were living in England 
and Wales (outside of the metropolis) just about half as 
many persons native to London as England and Wales 
had supplied to that city.** 

Thus we have here a case similar to that occurring so fre- 
quently in nature: on the same terrain where a more 
highly organized plant or animal has no longer room 
for subsistence, others less exacting in their demands take 
up their position and flourish. The coming of the new 
is in fact not infrequently the cause of the disappearance 
of those already there and of their withdrawal to more 
favourable surroundings. 

This process need not, however, in the world of human 
society necessarily be a process of displacement, a con- 
sequence of the imperfect equipment of the native ele- 
ments and of the superiority of the foreign ones. 

The reverse will perhaps occur quite as frequently, and 
in the examples cited is probably the rule. On account 
of the endless differentiation of labour in modern national 

" Comp. von Mayr, Statistik u, GeseUschaftslehre, II, pp. 122 fF. 
"London had in 1881, 3,816483 inhabitants. Of these there were 
born: 

D.*.««. P*"" cent, of the 

I'enoos. Population. 

In London 2^1,955 62.9 

In the immediate neighbourhood.. 384,871 lai 

Elsewhere in England and Wales.. 787,699 20.6 

In Scotland 49.554 13 

In Ireland 80,778 2.1 

In other countries 1 1 1,626 2.9 

On the other hand, 584,700 natives of London were counted in other 
parts of England and Wales. For every 100 persons from these terri- 
tories who had settled in London, 51 natives of London had left the 
metropolis. — According to the Ztschr. des preuss. statist. Bureaus, 
XXVI (1886), Statist. Correspondenz, p. xviii. 



CROfVTH OF TOIVNS CONSIDERED HISTORICALLY. 3^5 

economy it is the skilled labourers who experience most 
trouble in finding suitable employment and compensation 
for their labour where they live and have received their 
training, because it is there that the competition is keen- 
est. They emigjrate and seek more favourable surround- 
ings, better conditions of competition, while at these 
points less highly qualified labour may at the same time 
be in demand, which demand must be met by importation 
of labour from outside places. This less skilled labour 
may, on the other hand, however, form the stronger, bet- 
ter equipped element in its own locality; and though it 
may lack here the opportunity for a profitable utilization 
of its skill, its departure may, nevertheless, leave a void 
that it is impossible to fill. 

Thus the emigration of more highly trained technical 
labour from the cities was perhaps never greater than in 
the period of the so-called industrial boom of the seventies. 
At the same time, however, those same cities received an 
immense influx of labouring population from the country; 
and the departure of the latter again caused in the dis- 
tricts of great landed estates a serious dearth of agricul- 
tural labourers, an advance in wages, and in some places 
a lamentable condition of agriculture. Here, in every case, 
it was the relatively stronger that had emigrated, the rela- 
tively weaker that had remained; there could be no ques- 
tion at all of mutual displacement. 

Still less is there ground for such a view with regard to 
those internal migrations that have their origin not in 
the effort to find a better place for carrying on work, but 
in the search for more favourable conditions of living. The 
pensioned civil servant or military man who leaves the 
expensive metropolitan city for the country or a cheaper 
rural town ; the speculator who has become suddenJy rich 
and exchanged his fluctuating stocks for a solid country 



^ I 



3^6 INTERNAL MIGRATIONS OF POPULATION AND THE 

estate; the Parisian shopkeeper who enjoys his more la- 
boriously earned income in the quietude of his modest 
country cottage; and also, on the other hand, the Jewish 
cattle-dealer who has become wealthy and seeks the city 
in order to speculate on the exchange; Fritz Renter's ex- 
cellently portrayed Mechlenburg " Fetthammel " or rich 
farmer, who after disposing of his farm will enjoy the 
pleasures of city life; the poor clergyman's widow who 
moves into the city in order to give her children a better 
education and supplement her scant pension by keeping 
boarders; — none of these in their new places of residence 
enters into dangerous competition with the native labour- 
ing population. 

And yet at the objective points of the migration, even 
where the danger of displacement cannot enter into the 
question, there are innumerable struggles and endless fric- 
tion, all originating in the process of social amalgamation 
which is here always going on between the native popu- 
lation and the new-comers. The stranger has to adapt 
himself to his environment, to the peculiar local economic 
methods, to customs, speech and the political, religious 
and social institutions of his new abode. And the inhabi- 
tants of the latter place again, however settled in char- 
acter and peculiar in type cannot altogether escape 
the influences that rush in upon them from without. 
Though these influences often mean for them an increase 
of working energy, an expansion of the horizon, a breeze 
bringing freshness into corrupt local conditions, yet per- 
haps much more frequently they result in a loss of good 
old customs, of solid business qualities, of interest in the 
common weal, and, above all, of social characteristics. 

Now there can be no doubt that these struggles for 
mutual adaptation will take a vastly different form and 
course when waged between similar and between diver- 



GROIVTH OF TOIVNS CONSIDERED HISTORICALLY. 3^7 

gent elements. For this very reason the division used in 
municipal statistics for marking the distinction between 
native and resident population does not suffice for more 
exact socio-statistical investigations. 

For if, for example, it has been ascertained that the 
native-bom inhabitants of the city of Munich in 1890 
amounted to 36 per cent, of the whole, while in Hamburg 
they constituted 47.5 per cent., the mere fact that in the 
former city there are 11.5 per cent, more citizens of extra- 
mural birth is far from proving that the population of 
Munich is to this extent more heterogeneous than that of 
Hamburg, and that in the former the process of mutual 
social adaptation is attended with more violent friction 
and struggles than in the latter. In like manner the 
fact that two cities — for example, Dresden and Frankfurt- 
on-Main — show the same proportion of non-native to na- 
tive-born citizens does not mean that this process takes 
the same course in both. It is easy to conceive that the 
strangers in one city may show a greater homogeneity of 
customs and speech, economic energy and social habits 
amongst themselves and with the native population on 
account of coming from a neighbourhood more nearly 
akin, while in the other city heterogeneous elements from 
more distant localities are mingled together. 

The final result of the mutual adaptation of non-native 
and native population will be altogether different in each 
of these cases. While in the former individuals and groijps 
of persons of approximately like economic equipment and 
similar social character enjoy peacefully together the ex- 
isting conditions for business, in the latter perhaps the 
more robust, energetic, easily contented race will vanquish 
the decrepit, weaker and more pretentious in its ancient 
home, or at least eject it from the most favourable fields 
of industry. Especially is it true that a lower standard 



368 INTERNAL MIGRATIONS OF POPULATION AND THE 

of living can give the incomers a superiority over the na- 
tive labour in the competitive struggle, which involves 
the latter in the most deplorable consequences. The im- 
migration of the Polish labourers into the provinces on 
their west, of the Italians into Switzerland and south Ger- 
many, and of the Chinese into the cities of the North 
American Union are well-known examples of this. 

But even when the economic and social assimilation 
takes place without severe struggles there may persist be- 
tween incomers and natives differences that simply can- 
not be removed, invading and disturbing the original 
homogeneity of the population. We have especially 
in mind differences of creed, of language and of political 
allegiance. The two largest cities of Switzerland, Geneva 
and Basel, both of which we are accustomed to look upon 
as strongholds of Protestantism, have to-day, in conse- 
quence of influx from without, a population of which over 
a third is foreign. In Geneva about 20 per cent, of the 
population have a mother tongue other than French. 
Finally, since 1837 the Roman Catholics have increased in 
Basel from 15 up to 30 per cent, of the population, while 
in Geneva they have reached 42 per cent. Even he who 
has no detailed knowledge of the internal history of these 
small municipalities will be obliged to admit that such 
differences are not void of danger. 

If these considerations show that by no means the ma- 
jority of internal migrations find their objective point in 
the cities, they at the same time prove that the trend 
towards the great centres of population can in itself be 
looked upon as having an extensive social and economic 
importance. It produces an alteration in the distribution 
of population throughout the State; and at its originating 
and objective points it gives rise to difficulties which legis- 
lative and executive authority have hitherto laboured. 



GROIVTH OF TOIVNS CONSIDERED HISTORICALLY, 369 

usually with but very moderate success to overcome. It 
transfers large numbers of persons almost directly from a 
sphere of life where barter predominates into one where 
money and credit exchange prevail, thereby affecting the 
social conditions of life and the social customs of the 
manual labouring classes in a manner to fill the philan- 
thropist with grave anxiety. 

This mighty flow of the country population into the 
cities and the universally rapid rise of the latter in volume 
is looked upon by many as an entirely modern phenom- 
enon. In a certain sehse this is true. The eighteenth cen- 
tury knew nothing of it, at }east in Germany. The famous 
founder of population statistics, J. P. Siissmilch, did not 
succeed in discovering any regular law governing the 
movement of population in cities. He is of the opinion 
that they rise and fall in size according to the will of 
God.^^ J. H. G. von Justi deems it hardly possible that a 
city should increase unless special privileges be granted 
to the incoming settlers.^^ This is in accord with such 
population statistics as we have been able to collect for 
individual cities from the second half of the seventeenth 
century to about 1820;*^ these show retrogression and 
growth in irregular alternation. In France, on the other 
hand, the modern movement seems to have begun about 
one hundred and fifty years earlier ; and men already spoke 

" " Thus does the mighty ruler of the universe impart to states and 
cities might, riches, and glory. He takes from them again and gives 
to others according to his good will. He pulleth down the mighty 
from their seat and exalteth them of low degree."— (75«/»fA^ Ordnung, 
II, § 546 (2d ed., pp. 477, 478). 

*^ Grundsdtse d. Poliseiwiss., § 54. Comp. also GesatnmeUe polU, «. 
Finansschriften, III, pp. 449 fF. 

** Much material relating to the subject has been collected by Inama-, 
Sternegg in the Hand wort. d. Staatsw., II, pp. 433 if. 



\ 



370 INTERNAL MIGRATIONS OF POPULATION AND THE 

in the eighteenth century, according to familiar phrase- 
ology, of the " depopulation of the open country." ^ 

If on the other hand we go farther back into the history 
of man in Europe we find two periods showing the same 
phenomenon on a grand scale: ancient times, especially 
the era of the Roman Empire; the later Middle Ages, in 
particular, the fourteenth and fifteenth centuries. Between 
them lie great epochs of quiescence, if not of retrogression 
and decay. 

How are these earlier periods of migration to the cities 
to be regarded from the standpoint of the history of their 
evolution? Are they premature starts toward a goal 
whose attainment was reserved for our own time and its 
perfected means of communication? Or are they the out- 
flow of other impulses than those behind the correspond- 
ing movement of the present, and did they on that ac- 
count also lead to other results? Above all, was their 
influence upon population and their economic character 
the same? 

As concerns ancient times it would seem as if we must 
assume, in spite of the uncertainty of the population record 
handed down to us, that a consequence of the influx of the 
rural population was the inordinate growth of the cities.^* 
But it must not be overlooked that only a part of that 
population migrated of its own free will, namely, the free- 
men. The remaining and much larger portion, the slaves, 
were collected by their masters in the cities, or brought 
thither by the slave trade. 

*• Evidence collected by Legoyt, Du progrh des agglomerations ur- 
baines et V Emigration rurale (Marseilles, 1870), pp. 8 ff. 

"On what here follows compare particularly R. Pohlmann, Die 
Uehervblkerung d, antiken Grossstddte im Zusammenhange mit d. Gesamt- 
entwick. stddtischer Civilisation (Leipzig, 1884); also Roscher, System 
d. Volksw.f III, Introduction, and Biicher, Die Aufstdnde d. unfrei, Ar- 
better 143-129 v. Ckr. 



GROIVTH OF TOiVNS CONSIDERED HISTORICALLY. 371 

Where freemen moved in from the country they 
usually came not because a better prospect of economic 
advance in the cities beckoned them, but because they 
were deprived of their lands through the growth of the 
great slave estates. In the cities, it is true, they found 
all the lucrative branches of trade in the hands of slaves 
and freedmen; but they were here in less danger of starva- 
tion, inasmuch as the proletarian masses of the cities in 
whose midst they settled were supported by public and 
private largesses. 

The large cities of antiquity are essentially communities 
for consumption. They owe their size to the political cen- 
tralization which collected the surplus products of the ex- 
tensive areas cultivated by individual husbandry at one 
point where the governing class was domiciled. They are 
imperial, or at least provincial, capitals. Accordingly they 
first arise in the time of the successors of Alexander 
and reach their height under the Roman Empire. The 
capital, Rome, itself depends for its food-supply upon the 
taxes in kind from the provinces; and the same is later on 
true of Constantinople.^* It is a communistic and im- 
perialistic system of provisioning, such as the world has 
not seen a second time. The extortions of the officials, 
the farming of the revenues, the usurious practices, the 
g^eat estates of wealthy individuals worked by slaves, the 
state-recognised obligation to supply largesses of bread, 
meat and wine to the masses — ^all these placed the pro- 
ductive labour of half a world at the service of the capital 
city and left open to the private activity of its inhabitants 

" Krakauer, Das VerpHegungswesen d. Stadt Rom in d. spater. Kaiser^ 
Meit (Leipzig, 1874), and £. Gebhardt, Studien uberd, VerpHegungs- 
wesen von Rom u, KonstanHnopel in d. spater. Kaiser zeit (Dorpat, 1881). 
Also Rodbertus, Zur Gesch. d. rom. Tributsteuem in the Jhrb. f. N.-Oek. 
Oek. tt. Stat, VIII, especially pp. 400 if. 



372 INTERNAL MIGRATIONS OF POPULATION AND THE 

nothing but the sphere of personal services. From what 
we know of the larger provincial cities we may conclude 
that in them similar conditions prevail.*' 

A favourable market for free labour, a place for the 
skilled production of goods on a large scale for export, 
the ancient metropolitan city was not.*^ Anything resem- 
bling factory work rests, as does the extensive agricultural 
production, upon slave labour. Accordingly among the 
motives mentioned by the ancient writers as impelling the 
free rural population toward the cities the very one that 
is commonest to-day — ^the prospect of higher wages — 
plays no part. " Consider this body of people," writes 
Seneca*® to his mother; "the houses of the immense 
city are scarcely sufficient for them. From municipia and 
colonies, in short from the world over, have they come 
together. Some have been drawn hither by ambition, 
some have come on public business, others as envoys, 
others again have been attracted by luxurious tastes seek- 
ing an apt and ample field for indulgence, others by fond- 
ness for liberal studies, others by the shows; some have 
been led by friendship, others by enterprise, which here 
finds extended fields for displaying personal merit;** 
some have brought their personal beauty for sale, others 
their eloquence. There is no class of people which has 
not streamed to the city, where the prizes are great for 
virtue and vice alike." 

"*£. Kuhn, Die stddHsche m. burgerliche Verfassung d, Rom, Reichs, 
I> pp. 46 ff.» points to ah organization of the cura annona similar to that 
in the capital. 

"Francotte, Ulndustrie dans la Grhe ancienne, I, esp. pp. I49-I58» 
has now established this for the Greek cities. 

^Ad Helviam, 6. 

^Quosdam industria latam ostendenda virtuti nacta nuUeriam, It is 
competition that is meant, not " industry/' as Pohlmann, cited above, 
p. 17, translates it. 



GROIVTH OF TOIVNS CONSIDERED H!STORJCj4LLY. 37 J 

Quite different was it with the town-ward flow of popu- 
lation in the Middle Ages. Taken as a whole, it is per- 
haps not less voluminous than that at the time of the 
Roman Empire. It did not result, however, in the forma- 
tion of a few central points of consumption, but in the con- 
struction of a large number of fortified places distributed 
pretty evenly throughout the country, uniting within 
their walls all the organized industrial activity of the na- 
tion which was not attached to the soil. The mediaeval 
towns are originally mere places of refuge for the sur- 
rounding rural population ; ^^ their permanent inhabi- 
tants are the burghers, or people of the burg. Everything 
else — ^the market, the prosecution of trade, monetary 
dealings, the personal freedom of the town inhabitants 
and their special privileges before the law — ^is only a later 
consequence of this extra-mural military relationship. The 
defensive union became in course of time a territorially 
circumscribed economic union, for which the town or city 
was the trade centre and the seat of all specialized labour. 

The mediaeval cities'^ accordingly bear a great simi- 
larity to each other in the social and economic organizar 
tion of their population, and differ, as far as we can see,. 
only slightly in the number of their inhabitants. At their 
original founding the influx of the rural population 
seems often to have been far from voluntary. Later on 
the chief factor determining their growth was the greater 
security of person and property and the more varied op- 
portunities for earning a livelihood which they afforded 



••Comp. above, pp. ii6flF. 

" That is, so far as they really deserve the name. It is a peculiar in- 
consistency to attempt to-day to demonstrate the character of the 
mediaeval city by taking as examples places which never arrived at a 
true city status and which can bring forward no better claim to the 
name of city than that they were endowed with city privileges. 



374 INTERNAL MIGRATIONS OF POPULATION AND THE 

landless freemen and serfs. Their whole development, 
economically and numerically, came to an end, however, 
the moment all the handicrafts that the limited extent of 
the city-market areas was capable of sustaining were rep- 
resented and supplied with a sufficient number of master- 
workmen. Up to this point the cities offered complete 
freedom of movement and almost unimpeded access to 
g^ild privileges and burgess rights, while the rural land- 
owners, on the other hand, sought through limitations of 
the right of removal to secure themselves against the loss 
of their serfs. When, however, the cities were able to sup- 
ply all branches of trade from the internal g^rowth of their 
population, they also exhibited a willingness to check ac- 
cessions from without, and hence brought about those 
numerous obstacles to settlement and to entry upon a 
trade which have persisted into modern times. There arose 
a sharp division between city and country. Migration to 
and fro naturally continued to a certain extent, but it was 
confined in the main to an exchange of labourers among 
the cities themselves. City development had fallen, as it 
were, into a condition of numbness from which it could be 
roused only through transition to a new economic order. 
We are in a position to prove statistically for a few 
localities the statement just made. There have been in- 
stituted exhaustive investigations into the origin of the 
-mediaeval population of Frankfurt-on-Main,'* and re- 
'Cently also regarding certain sections of the population 
♦of Cologne.®' From these it appears that the majority 
*of the persons received by these two cities as burghers 

" Bucher, Bevblkerung von Fr,, pp. 163 flF., 304 flF., 422 ff., 521 ff.. 591 ff., 

627 ff. 

" A. Doren, Unterswhungen s, Gesch. d, Kaufmannsgiiden d. Mittclaiters 
<in Schmoller's Forschungen, XII, 2), Appendix 1; and now also H. 
Bunger's Beitrage z, mittelalt. Topograph,, Recktsgeschick u. SosiaUtatisttk, 
d. Stadt Koln (Leipzig, i8g6), Sec. 3. 



GROIVTH OF TO^NS CONSIDERED HISTORJCj4LLY, 375 

during the fourteenth and fifteenth centuries migrated 
from the country. Of every lOO new burghers there 
came to: 

In the Period. From Qtie.. ^"^H^JSSf "* 

Cologne 1356- 1479 37-4 62.6 

Frankfurt 1311-1400 28.2 71.8 

" 1401-1500 43.9 56.1 

We see from this that in the last two centuries of the 
Middle Ages the movement of population from the coun- 
try to the cities, though it continued, was on the wane, 
while the admixture of town elements among the new 
burghers increased. Thus as early as the fifteenth century 
certain strata of the population of Frankfurt received their 
chief increment through emigration from other cities. Of 
the incoming Jews, for example, 90 per cent., and of the 
members of a fraternity of journeymen metal-workers 79.3 
per cent., came from cities. The material from which the 
last percentage is deduced also covers, it should be said, 
the first quarter of the sixteenth century. 

Unfortunately, further figures regarding the sixteenth 
and seventeenth centuries are not available. But for the 
period from the beginning of the eighteenth till after the 
middle of the nineteenth century we can offer some fig- 
ures which serve to show that there was an epoch when 
the urban handicrafts received their workers almost ex- 
clusively from other towns. The Frankfurt municipal 
archives contain a number of books regarding the lodg- 
ing-places of the bookbinders, in which are recorded the 
names and places of origin of all the journeymen of this 
craft who came to Frankfurt between 171 2 and 1867 
(14,342 person^ in all). Some years ago we worked over 
this extremely valuable statistical material and found that 
of every 100 incoming journeymen bookbinders there 
came: 



376 INTERNAL MIGRATIONS OF POPULATION AND THE 

Period.. Fromaties. ^^h^^SS!*"* 

1712-1750 97.5 2.5 

I75I-I800 94.3 5.7 

I80I-I835 89.2 10.8 

I836-I850 86.0 Z4.0 

1851-1867 8i.a 18.8 

We see here how, in a trade of a specifically urban char- 
acter, within a period of rather more than a century and 
a half, the proportion of workers drawn from the country 
has continuously increased. Had it been possible to con- 
tinue the investigation for the period from 1867 down to 
the present time, we should undoubtedly have found that 
the balance has inclined more and more in favour of the 
journeymen from rural localities. 

In the contemporary migrations to the cities a fusion 
of town and country strongly resembling that established 
by us for the fifteenth century seems to have set in.'* 
Of every 100 of the inhabitants born in other places there 
were in: 

Year. Of City Birth. Of Country Birth. 

Leipzig 1885 50.6 49.4 

Basel 1888 ^ 23.5 76.S 

As in the Middle Ages, the city element relatively in- 
creases and the country element decreases according to 
the distance of place of birth from place of settlement. The 
various classes of the population show but slight differ- 
ences in this regard. Generally speaking those occupa- 
tions that demand a special training have a stronger ad- 

** Only the simplest results of these investigations can be given here. 
Details may be found in my Bevblkerung d. Kantans Basel-Stadt am i. 
DeM. j888, pp. 62 ff. We may also refer to Hasse*s Ergebnisse d, 
Volkszahlung vom i. Des. 1885 in der Stadt Leipzig, Pt. II, pp. 7 ff. The 
higher figures in the rural accessions for Basel are explained by the 
fact that in the above work the city limits are made to include only 
3,000 inhabitants. 



GROWTH OF TOH^NS CONSIDERED HISTORICALLY. 377 

mixture of city elements than the spheres of simple manual 
labour. 

It is greatly to be regretted that similar statistical in- 
vestigations have not been carried out for a larger number 
of modem cities. From the evidence at present to hand 
we are apparently driven to conclude that the number 
of incomers of city origin is relatively greater in the large 
cities than in the medium-sized and smaller ones.^® The 
explanation of this phenomenon is a simple one. A large 
city exercises upon the population of the smaller cities the 
same power of attraction that the latter have for the pop- 
ulation of the country. In this way the transitions from 
one social and economic sphere to another are rendered 
less violent. Thus a gradual elevation of the migrating 
masses takes place, as also from generation to genera- 
tion a continuous preparation for the demands of life in a 
great city, which must render less violent the conflicts 
inevitable to the process of mutual adaptation within the 
new sphere. 

But if the cities of to-day exhibit a process of redis- 
tribution of population similar to that of their mediaeval 
prototypes, the resemblance is only superficial. In the 
fourteenth and fifteenth centuries we have to do with the 
last stages of an evolution whose ultimate result was the 
formation of numerous .small autonomous spheres of 
economic activity, each of which exactly resembled the 
other in its harmonious development of production; in 
the nineteenth century we have to deal with an increas- 
ing differentiation of the individual centres of population, 

"Besides the work on Leipsig already mentioned, a later exhaustive 
treatise on the accessions to, and losses of, population in Frankfurt-on- 
Main in the year 1891, published by Dr. Bleicher in the Beitr, i. 
Statistik d. St. Frkf., II, pp. 29 ff., gives interesting information re- 
garding this point. 



37^ INTERNAL MIGRATIONS OF POPULATION AND THE 

in accord with the designs of a greater whole, namely, of 
a state-regulated national economy. 

This process begins with the development of the mod- 
ern State and modern national administration. Hitherto 
each city had developed within itself all the branches 
of city life not forbidden by local conditions; now one 
city becomes a permanent royal residence, others be- 
come seats of district and provincial administrations, of 
prisons, of higher educational institutions and of all kinds 
of special branches of administration, while still others 
become garrison cities, border fortresses, fair-towns, 
watering-places, junction-points of commercial routes, 
etc. They take over definite functions for the whole coun- 
try and for all other places, though these functions are 
not always specifically urban. The cities may also form 
alliances with rural residence centres. This process has 
been especially prominent since the fuller development 
of city industry on a large scale and the extraordi- 
nary increase and perfection of the means of communica- 
tion. In this new national era the total national produc- 
tion endeavours so to distribute itself over the territory 
controlled by it that each of its branches may find the 
location best suited to it. Factory and house-industry 
districts arise, and separate valleys and whole regions take 
on a semi-urban character. Certain cities develop special 
branches of industry and trade reaching out far beyond 
the local, and often even the national, demand. In others, 
again, all industry and business life decline; they sink 
down to the level of villages, so that the historical rights 
of burgess that still attach to their name appear in striking 
contrast with their position as places of trade and with 
the number of their inhabitants. The distinctions between 
city and country are blotted out. This happens in the 
neighbourhood of rising industrial cities through the 



\ 



GROIVTH OF TOIVNS CONSIDERED HISTORICALLY. 379 

planting of factories and workmen's dwellings in the sub- 
urbs and beyond; in the neighbourhood of the declining 
" rural cities " through the approach of the latter to the 
condition of surrounding country places, and through the 
rise of populous industrial towns. On the whole, how- 
ever, the number of centres of population and of objective 
points for internal migrations is to-day relatively much 
smaller than in the second half of the Middle Ages.^' 

But in still another respect does the redistribution of 
population resulting from the internal migrations of the 
present time differ from that witnessed by our ancestors 
from the tenth to the fifteenth centuries. In consequence 
of the gfreater certainty of a living and of far-reaching 
measures for the health of the people the increase in pop- 
ulation is to-day more rapid than in mediaeval times. It 

••The German Empire had in 1890 a total of 2,285 "cities/* Of 
these there were 26 with more than 100,000 inhabitants, 22 with from 
50,000 to 100,000, 104 with from 20,000 to 50,000, and 169 with between 
10,000 and 20,000. Besides these there were 56 villages and suburban 
municipalities with from 10,000 to 50,000 inhabitants, 11 of them with 
more than 20,000. — In Prussia there were in that year 46 " cities " with 
less than 1,000 inhabitants, 14 of these being in the Province of Posen, 
12 in Silesia, 10 in Hesse-Nassau, 3 in Brandenburg, 2 each in West 
Prussia and - Westphalia, i each in Saxony, Hanover and the Rhine- 
land (Schleiden with 515 inhabitants). Alongside these dwarflike 
cities there were 37 rural municipalities with more than 10,000 inhabi- 
tants. — How far some of the old cities have declined is shown by the 
following fignires for the Grand Duchy of Baden. There the census 
of 1885 gave 114 "cities," only 6^ of these having a population of over 
2,000, and 9 with over 10,000. Of the remaining 51 " cities " 42 had from 
1,000 to 2,000 inhabitants, 4 had from 500 to 1,000, and 5 below 500 
(among these last being Kleinlaufenburg with 441, Neufreistett with 
427, Blumenfeld with 349, Furstenberg with 341, Hauenstein with 157). 
For every city there were on an average 14 villages. On the other 
hand, there were altogether 129 municipalities with over 2,000 inhabi- 
tants, 66 of these being " villages." Of the old cities only 55 are thus 
cities according to the modern idea, and of the villages four per cent 
are from the point of view of population to be reckoned in with the 
cities. 



33o INTERNAL MIGRATIONS OF POPULATION AND THE 

is safe from those heavy reverses so frequently resulting 
in those ages from harvest failure, feuds and plague. 
On that account the modern migrations into the large 
cities and industrial districts in many cases absorb only 
a surplus population that would not find sufficient room 
for earning a livelihood in the places of its origin. 
At these points they retard or completely check the con- 
gestion of population; while on the other hand at the 
points of agglomeration no economic obstacles bar the 
way to a continuous and rapid increase. 

In mediaeval times, on the contrary, the migratory ac- 
cession of population was distributed among a multitude 
of walled places scattered at fixed intervals over the 
whole country. The increase in many cases continued 
only until the city was full. When once it had as many 
inhabitants as it needed to man its walls and towers and 
supply all the branches of industry, there was no room for 
more. Extensions of the city limits often did take place 
in mediaeval times, it is true; they are the result of the 
increasing formation and subdivision of special trades. 
But the Middle Ages developed no large cities; the 
mediaeval economic and commercial system forbade it. 
The country was often deprived of the population neces- 
sary for the cultivation of the soil; yet even with such 
accessions the frequency of extensive losses kept the city 
populations stationary. 

From these remarks it will indeed remain uncertain 
whether or not the internal migrations that accompanied 
the development of the industrial life of the mediaeval town 
were relatively more extensive than the corresponding 
territorial movements and shiftings of population that 
result to-day from the more national character of settle- 
ments. On the other hand, it cannot be doubted that the 
attraction of the great cities of modem times for the pop- 



GROWTH OF TOIVNS CONSIDERED HISTOIUCALLY. 381 



ulation of the smaller towns and the country is exerted 
over greater expanses of territory than the mediaeval towns 
held within the circle of their influence.. One is not in a 
position to say, however, that the recruiting territory for 
the population of a city has expanded since the begin- 
ning of modern times in direct ratio to the number of its 
inhabitants. On the contrary, one is astonished to find 
what a slight effect the perfecting of the means of com- 
munication and the introduction of freedom of movement 
have had upon the extent of the territory covered by the 
regular internal migrations. 

A few figures will make this clear. Of every hundred 
new settlers coming to Frankfurt, Oldenburg and Basel 
the numbers according to distances are as follows: 



City. 



Frankfurt . . 
I* 

Oldenburg. . 

Basel 

II 
II 



Class of Population. 



New citizens 

ii II 

Metal workers 

Citizens born in other 
places 

Citizens born in other 
places 

Journeymen craftsmen.. 

Factory laborers 



Period. 



14th century 
15th 

15th and i6th 
centuries 



1880 

1888 
1888 
1888 



0-9 

Miles. 


N?iiil. 


46.7 
23.1 


39.3 
52.7 


2.7 


45.0 


21.8 


42.1 


16.7 

13.9 
17. 1 


50.2 
40.0 
59.6 



Over 45 
Miles. 



14.0 
24.2 

52.3 

36.1 

33- 1 
46. z 

23.3 



Of the three recruiting zones distinguished here the 
outermost at present contributes more to the total popu- 
lation and the inner less than in mediaeval times. The 
reason probably is that to-day the population in the 
more immediate neighbourhood of a city takes advantage 
of the city's labour market without settling in the city 
itself, whether it be that they go daily to their places of 
work in the city by special workmen's trains or other con- 
venient means of transportation, or that the great indus- 
tries of the towns erect their workshops in neighbouring 



3^2 INTERNAL MIGRATIONS OF POPULATION AND THE 

places. The recruiting territory for journeymen has 
rather contracted as compared with mediaeval times. 
With this is linked the circumstance that at present three- 
fourths of this class of workmen are drawn from the coun- 
try, while at the close of the Middle Ages less than one- 
quarter of them came from villages and hamlets. Of the 
Frankfurt journeymen metal-workers in the fifteenth and 
sixteenth centuries only 20.7 per cent, were born in the 
country; of the Basel bakers and butchers in 1888, on the 
contrary, 78.7 per cent, and of the journeymen of other 
handicrafts 75.2 per cent, were of country birth. Even 
to-day journeymen craftsmen still migrate in much larger 
numbers and to greater distances than the typical work- 
man class of the present, the factory hands. Of the 
factory workingtnen in Basel in 1888, 25.8 per cent, were 
born in the city itself; of, the handicraft journeymen, only 
16.3 per cent. How many of them were bom and still 
domiciled in the immediate neighbourhood the statistics 
unfortunately do not show. But all modem industrial 
development tends in the direction of producing a per- 
manent labouring class, which through the custom of 
early marriage is already much less mobile than the jour- 
neymen of the early handicrafts, and which in future will 
doubtless be as firmly attached to the factory as were the 
servile labourers of the mediaeval manor to the glebe.®'' 
If this is not very noticeable at present it is because the 
majority of large industries have not yet attained their 
growth, and because it is necessary for them, as long as 
they extend their works, to meet the increased demand 

The construction of workingmen's dwellings by the great man- 
ufacturing establishments, whether these pass over into the possession 
of the labourers or are rented to them, is even now begetting a sort of 
factory bondage, which has an appalling resemblance to the old bond- 
age to the soil. Comp. my article on the Belgian social legislation in 
Braun's Archiv. f. soz. Gesetzg. u. Stat, IV, pp. 484, 485. 



GROIVTH OF TOH^NS CONSIDERED HISTORICALLY. 383 

for labourers by drawing further upon the surplus popu- 
lation of the rural districts. 

These remarks point to the conclusion that we are not 
justified in attributing a growing migratory character to 
society as a result of the closer network of commercial 
routes and the invention of perfected means of communi- 
cation. Rather should we say that at present we are in 
the midst of a transition period in which the yet un- 
completed transformation of the town and territorial 
economic structure into a national one involves a con- 
tinuous displacement of the boundaries of division of 
labour and of the centres of the various branches of pro- 
duction, and consequently displacements of the labouring 
population. 

After a period of economic and social ossification ex- 
tending over centuries, in which all sorts of limitations 
upon migration and settlement held the population fast 
to the original ancestral seats, it is not surprising that 
many view with anxiety the great movements of popu- 
lation which to-day extend over whole territories. It 
seems almost as if the early times of universal migration 
were returning. But in this they forget that it is only a 
part — the rural part — of the population that has become 
more migratory; that up to the early years of last 
century a great number of these were bound to the soil. 
The merchant, the craftsman and the scholar are to-day 
much less mobile than, for instance, in the time of the 
Reformation; and the industrial labourers move relatively 
less often and to shorter distances than they did even a 
century ago. Only their number has become much greater 
and is still steadily increasing. This growth o\ in- 
dustry displaces a part of the rural labouring population 
from their usual places of abode, to which nothing now 
holds them but the interests of those who profit by their 



384 INTERNAL MIGRATIONS OF POPULATION AND THE 

helplessness. The further progress of this movement will 
probably show, even in a few decades, that the human 
race in the course of its evolution has on the whole be- 
come more stationary. 

We may thus say in conclusion: In this general influx 
to the cities and their suburbs we are to-day undergoing 
what our ancestors experienced in the second half 
of the Middle Ages, the transition to a new economic 
and social order and a fresh distribution of population. 
At that time the movement inaugurated the period of 
town economy and of sharp separation of town and 
country; the movement in the midst of which we now 
live is the outward sign that we have entered upon a new 
period of development, — the period of organic distribu- 
tion of population, the period of national division of 
labour and of national satisfaction of wants, in which the 
distinctions between city and country as places of abode 
are being obliterated by numerous transitional forma- 
tions. This fact has long since been recognised by statis- 
ticians who have dropped the historico-juridical concep- 
tion of city and substituted a statistical one in which 
places are distinguished according to the number of their 
inhabitants. 

Every transitional epoch has its inconveniences and its 
suffering. But the modern movement of population, in 
so far as it is expressed in the influx into the cities, will, 
like that of mediaeval times, reach its goal and then sub- 
side. This goal can be none other than to assign to 
every individual capacity and to every local gproup of per- 
sons that place and that role in the great national life 
in which its endowments and the altered technical con- 
ditions of economic activity best fit it to contribute to 
the general welfare. 

Thus froni a consideration of internal migrations, dc- 



GROWTH OF TOlVhIS CONSIDERED HISTORICALLY. 385 

Spite the fact that the conditions accompanying them 
are often far from pleasant to contemplate, we may gain 
the assurance that Jhey too mean, from the wider stand- 
point, an advance towards higher and better forms of 
social existence, both for the individual and for the race. 



y